(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Identification of anti-idiotypic antibody during an immune response in dogs"

THE IDEOTIFICATION OF Al'JTI-IDIOrYPIC ANTIBODY 
DURING AN liVlMUNE RESPONSE IN DOGS 



BY 



KEVIN T. SCHULTZ 



A DISSERTATION PRESENTED TO TEE GRADUATE OOUNCIL 
OF THE UNI^\7ERSITY OF FLORIDA IN 
PARTIAL FULFILLMENT OF THE REQUIREMENTS 
FOR THE DEGREE OF DOCrOR OF PHILOSOPHY 

UNIVERSITY OF FLORIDA 



1983 



This dissertation is dedicated to my wife Nancy. 
Without her love and support (and typing), it 
would never have been. 



ACKNCWLEDGEMEOTS 



I wish to express ir^ deep appreciation and thanks to 
some very important people in mv life. First of all, to 
parents, m/ wife and ny family for their constant love and 
support . 

Very special thanks are extended to Dr. Richard 
Halliwell for liis help, guidance and friendship throughout 
iTY graduate experience. 

To Drs. Gail Kunkle, Robert Mason and Bill Hollonan, 
ny deep appreciation for their unwavering friendship (and 
occasional needed prods) along ray journey. 

I also wish to thank Drs. K.I. Berns, M.D.P. Boyle, 
R.B. Crandall, A. P. Gee, G.E. Gifford, M.J. P. Lawman, and 
P. A. Small, Jr., for their suggestions, help and 
encouragement , 

Last (but not least), I would also like to thank m/ 
fellow graduate students and all my friends for mking this 
experience a memorable one. 



TABLE OF QDNTENTS 



Page 

ACKISOWLEDGEMENTS iii 

ABSTRACT... vi 

OlAPrER ONE INTRODiXTION 1 

OiAPrER 'HAfO THE INDUCTION AND KINETICS Of AN 
ANTI-DNP IGE RESPONSE 

Introduction 17 

f'laterials and Methods 18 

Results 31 

Discussion 55 

Sunmary 57 

Conclusions 57 

OiAPTSR THREE ATTEMPTS TO REGULATE AN ANTI.BODY 
RESPONSE WITH AUTOLOGOUS ANT'IEODY 

Introduction 59 

Materials and Methods 60 

Results 62 

Discussion 74 

Summary and Conclusions 76 

CHAPTER FOUR THE IDENTIFICATION OF ANTI- 
IDIOTYPIC ANTIBODY 

Introduction 77 

Materials and Methods 78 

Results 82 

Discussion 94 

Summary 99 

Conclusions 99 

CHAPTER FIVE DETECTION OF ANTI- IDIOTYPIC 
.ANTIBODY USI.NB AUIOLOGOUS ANTI-DNP 
F(AB)' FRAGMENTS AS THE IDIOTYPIC 
ANTIGEN 

Introduction 99 

Material and Methods 100 

Results 103 

Discussion • 130 

Summary and Conclusions 135 

iv 



CHAPTER SIX CD^XiUSIONS - 137 

EffiFERENCES 140 

BIOGRAPHICAL SKETCH 149 



V 



Abstract of Dissertation Presented to the Graduate 
Council of the University of Florida 
In Partial Fulfillment of the 
Requiranents for the Degree of Doctor of Philosophy 

THE IDENTIFICATrON OF AIvlTI- IDIOTYPIC ANTIBODY 
DURIiNG AlSI Iimum RESRDNSE IN THE DOG 

By 

Kevin T. Schultz 
August, 1983 
Chairitan: Richard E. Halliwell 

Major Department: Immunology and Medical ivlicrobiology 

This investigation ooramensed with the development of 
an aninal model to study the synthesis of IgE antibody. 
Repeated iomunizations with a haptenated parasite extract 
(dinitrophenol coupled ascaris) in young dogs resulted in 
the production of anti-DNP antibody of the IgE, IgG and IgM 
class . 

Although attempts to regulate this anti-hapten anti- 
body response by administration of autologous anti-DNP anti- 
body were unsuccessful, such tlierapy did result in the pro- 
duction of ant i- idiotypic antibodies. These ant i- idiotypic 
antibDdies were demonstrable using mouse hybridoma-derived 
anti-DNP antibodies. 



vi 



This antibody was shown to be anti-idiotypic rather 
than an internal image of antigen because it bound to only 
bMD of four monoclonal anti-DfJP antibodies and failed to 
inhibit the id anti-id interaction with hapten. Anti- 
idiotypic antibodies were detected during the immunization 
schedule in three of five dogs using autologous anti-DNP 
F(ab)'2 fragments as the source of idiotypes. 

The anti-idiotypic antibodies identified using the 
mouse monoclonal antibody were tlie result of the immuni- 
zation procedure and did not appear to be physiologically 
relevant to regulation of the immune response. On the other 
hand, the anti-idiotypic antibodies identified with the 
autologous source of idiotypes appear to be produced during 
tlie DNP-ASC immune response and were detected before autol- 
ogous antibody iinmunization. The antigens that induced the 
anti-idiotypic response appeared to be, in tiiis case, the 
idiotypes on the anti-DNP antibody that were produced frcm 
tiie DNP-ASC immunization. 



vii 



CHAPTER ONE 
INI'RODUCriON 

Allergic diseases of the immediate type are very 
important pathologic disorders in both iran and dogs. 
Clinical signs are initiated by an interaction of antigen 
and IgE antibody with resultant inediator release fron nast 
cells and basophils. In iran and dogs, allergic reactions 
cause considerable morbidity and can be fatal (1,2). 
Anaphylaxis fron a bee sting is a classic exainple for both 
species . 

Atopy in man is an inherited disease which is 
associated with an antigen specific IgE response against 
environirental allergens. This disease is expressed 
clinically as asthma, hay fever, atopic dermatitis or any 
conbination of these three (3). The dog is an excellent 
experimental animal model to study IgE n^iated hyper- 
sensitivity for atopic diseases of mn because of the 
similarity of the allergic reaction in both species (1,2,4). 
Canine IgE shares many E^ysicochemical properties \'n.th 
human IgE (5,6,7). Dogs like rren, develop spontaneous 
disease associated with increased synthesis of IgE antibody 
(3,4) and in dogs the disease is also familial (4). 



1 



2 



There are a number of unique features of IgE antibody 
synthesis. Firstly, IgE circulates in very sitbII anraunts as 
compared to other antibody classes. In man, the serum level 
of this antibody is about 1/10,000 the level of serum IgG 
(3), and in dogs serum IgE is about 1/100 the level of serum 
IgG (7). Serum IgE levels of internally parasitized people 
and dogs are elevated as compared to non-parasitized indivi- 
duals (7,8,9). 'The higher IgE level in dogs is felt to be 
the result of a greater parasite burden in this species (7). 

Secondly, IgE is produced predominantly locally by 
lynph nodes in the respiratory and gastrointestinal tracts 
as well as in regional lym^i nodes (10). These observa- 
tions have led to the suggestion that this iiraiunoglobulin is 
important in host defense of mucosal surfaces and partic- 
ularly against parasites. Furthermore, IgE has been shown 
to participate in parasite killing through antibody depen- 
dent cell mediated cytotoxicity (11). Thirdly, the antigens 
that stimulate IgE antibody are usually very complex and 
heterogenous substances such as allergens or parasites and 
their extracts. When an animal is exposed to these anti- 
gens, the antibody response usually includes high titer IgE 
antibody v\*iereas bacteria and viruses usually do not induce 
IgE antibody in spite of being very immunogenic (12). 



3 

The induction of IgE antibody experimentally requires 
special conditions. For exarrple, high doses of antigen and 
strong adjuvants such as complete Freund's adjuvant are 
unfavorable to the developiiient of an IgE response >^ereas 
low doses of antigen in an adjuvant such as aluminum 
hydroxide tend to favor IgE production (13). Furtherrtore, 
if haptens are coupled to parasite extracts, high titer 
anti-hapten IgE antibody responses will frequently develop. 
However, if the same hapten is coipled to a T- independent 
antigen or a different T-dependent carrier there is usually 
no IgE response (13,14). 'This suggests that IgE production 
is dependent on both the carrier and T-cells. The reason 
why parasites and their extracts are efficient inducers of 
IgE antibody is not fully understood. It is known tl^at this 
eniiancing effect is modulated tlirough factors produced by T 
cells. Ishizaka's group (15) have shown tliat T cells 
derived frcm (N. brasiliensis ) parasitized rats produce 
an IgE-potentiating factor which selectively potentiates a 
non-specific IgE antibody response. This factor has 
affinity for IgE, binds to IgE-bearing B cells tiirough 
surface IgE and enhances tlie differentiation of these cells 
into IgE forming cells. A factor with similar properties 
has been produced from T- cells obtained fran patients with 
hyper- IgE syndrome, suggesting that the regulatory factors 



4 

and pathways for enhancing igE antibody production for 
parasites and IgE in general might be similar (16). 

A number of approaches have been used in an attempt to 
control allergic disease. Avoidance of the antigen is one 
approach, but this is rarely jx^ssible. Drugs tliat inhibit 
mediator release or control the effects of that release are 
also anployed, but they have side effects and often require 
continual therapy. However, the ideal approach would be to 
regulate the production of the unwanted IgE antibody. 

The mechanisms used to regulate IgE antibody responses 
involve either the inactivation of B cell precursors or tlie 
mnipulation of T cell populations. To tliis end, primary 
and ongoing antibody responses, including IgE antibody, have 
been suppressed in mice by antigen coupled to non- immuno- 
genic carriers such as d-glutamine-d-lysine (dGL) or poly- 
vinyl alcohol (17,18). Both of these carriers inactivate 
hapten- specific B cells and can induce hapten-specif ic 
suppressor T cells, r-loreover, when dGL is coupled to 
proteins rather than haptens, the resultant suppression in 
mice is isotype specific (i.e. suppresses IgE alone) (19). 
Unfortunately, there are no published results of tlie use of 
this compound in dogs or man. 

Hyposensitization has also teen used in an effort to 
control allergies in both man and dogs ( 20-22 ) , The 
mechanism by which it works is not clear. It is knovm that 



5 

IgG antibody can have a role in regulating allergic syirptans 
( 22 ) . An IgG response can be induced by administration of 
allergen either by the normal route of exposure or by a 
route other than for normal exposure (i.e. subcutaneous 
versus inhalation). This IgG antibody presumably completes 
the allergen- IgE antibody interactions (21). However, this 
therapy is not without side effects (20). Furthertrore , only 
about 65 per cent of patients treated with hyposensitization 
have clinical improvement (20). 

An alternate type of hyposensitization involves 
modifying the allergen, usually by mild denaturation . 
Studies in mice with urea denatured ragweed showed that such 
treatment reduced allergenicity \^^ile maintaining immuno- 
genicity of the allergen. If large doses of urea denatured 
ragweed were given to mice previously sensitized to 
unmodified ragweed, such therapy resulted in antigen- 
specific T suppressor cell induction without the development 
of anaphylaxis ( 12 ) . These cells suppressed the anti- 
ragweed IgE response. A controlled study is underway to 
determine if this form of immunotherapy is any more 
effective in controlling allergic symptons than conventional 
hyposensitization . 

Another approach is to regulate the response with 
products of the immune systen. Smith (23), in 1909, was the 
first person to recognize tliat antibody could suppress tlie 



6 

developnent of an inmune response. In t±iese experiments he 
showed that certain mixtures of diphtheria toxin and anti- 
toxin oould be very iixmunogenic in guinea pigs, but if there 
was a large excess of antitoxin, the immunized guinea pig 
wauld fail to mount an immune response against the toxin. 
Numerous studies in the 1950 's and 1960 's verified this 
observation and also demonstrated that the isotype, amount, 
affinity and time of administration were important variables 
in determining the degree of suppression tliat passive anti- 
body had on the immune response (reviewed in 24). For 
example, IgG antibody given after antigenic exposure was 
more effective in inducing antibody suppression than IgM 
antibody. Further, the suppressed state was longer lived 
using IgG than IgM antibody. An interesting report by Chan 
and Sinclair (25) stated that the administration of anti- 
SR3C antibody given to mice after antigenic challenge led to 
a suppression of this response and tliis tolerant state could 
be transferred from one mouse to another with T-cells frcm 
the tolerized mouse. They suggested that the regulatory 
action of antibody operated through sane sort of "induced 
pathway or secondary immune response" (25 p. 977). 

In the ^rly 1970 's it was likewise shown tliat IgE 
antibody could be regulated by passively administered 
antibody (26-28). Rabbits were immunized to produce high 
titer IgE antibody and were given passive antibody 24 hours 



7 

after antigenic (diallenge. A cx>mplete inhibition of the 
passive cutaneoias anaphylaxis titer and a mrked decrease in 
the hemagglutination titer of these rabbits resulted as 
compared to controls ( 26 ) . It vas shown by- Tada and Okumura 
(27) that, in the rat, the administration of anti-DNP 
ascaris antibody resulted in mrked suppression of a 
preexisting IgE antibody response and this suppression was 
mintained for an extended period of time. This was in 
contrast to studies in the mouse in which administration of 
anti-ovalbumin IgG had little effect on the preexisting 
anti-ovalbumin IgE response { 28 ) . These differences were 
explained as species variation. Alternatively, they may be 
due to tlie difference in the antigenic systan employed. 

One explanation for the mechanism of regulation by 
passive antibody is tiiat tlie administration of this antibody 
acted as an antigen and stimulated an anti-antibody 
response. Lahss et al. (29) were tiie first to show that 
sane anti-antibodies WDuld bind to structures on antibody 
close to or within tlie antigen combining site. These deter- 
minants have been named idiotypes (id) and the immune 
response directed to tliem is tem^ an anti-idiotypic 
(anti-id) response. In 1974, Jerne (30) proposed his network 
hypothesis of antibody regulation. The basic premise of 
this theory is tiiat the immune system is regulated by a 
network of interactions between id and anti-id. A number of 



8 

assimptions are crucial premises to this theory. Firstly, 
most idiotypes exist at a level too low to induce tolerance. 
Thus, antigenic stimulation and expansion of these id will 
stimulate the production of a reciprocal set of anti-id. 
The id is then regulated directly by the ant i- id, indirectly 
by the anti-id on T-cellsor or by anti-id acting on T-cells. 
As the concentration of anti-id reaches sane critical 
threshold, a second anti-id response develops vvhich is 
specific for the id of the anti-id. This anti-id wDuld, 
therefore, be an anti-(anti-id) and would then stimulate a 
fourth response and so on, thereby resulting in an inter- 
related network of regulation between antibody molecules. 
Jerne also stated that id determinants can be present not 
only on antibody molecules of one specificity, but way be 
present on unrelated antibody molecules. Thus, antibody 
against antigen x might share some ids with antibody against 
antigen y. Lastly, although anti-id usually suppresses the 
corresponding id, it can be stimulating for tlie id as well. 
The anti-id viculd be expected to have a three dimensional 
structure similar or identical to the specific antigenic 
determinant. This type of anti-id is termed an internal 
image of antigen. 

,The- characteristics of idiotypes of antibody molecules 
have been described (31-35). In many instances, idiotypes 
are located in or very near to the antigen binding site. 



9 

This has been denranstrated by hapten inhibition studies. 
Brient and Nisonoff (31) induced anti-p-azobenzoate anti- 
bodies in rabbits. These antibodies were purified and 
injected into allotypically natched rabbits and the resul- 
tant antiserum bound to determinants present on some rabbit 
anti-p-azotenzoate antibodies. They then studied the 
effects that adding increasing concentration of hapten would 
have on the reaction between radiolabelled anti-azobenzoate 
antibodies and the ant i- idiotypic antiserum. They found 
that the binding affinities of the benzoate derivatives 
correlated closely with their ability to inhibit the 
ant i body/ant i- id interaction. In many otlier studies 
(32-34), anti-id was induced in anirrals immunized with an 
anti-hapten antibody. This anti-id was purified from the 
sera by initial adsorption to an affinity column having the 
iixmunizing antibody bound to it and was then eluted with tlie 
appropriate hapten. This purification process tlien would 
select for ant i- idiotypic antibodies which were directed to 
those idiotypic determinants very close to or within the 
antigen binding site and it would be expected that hapten 
could inhibit the id/anti-id interaction. 

On the other hand, it is not always possible for hapten 
to inhibit id/ant i- id. For example, Sher and Cohn (35) 
showed that there was variation in the ability of hapten to 
inhibit id/anti-id interaction. Hapten was not able to 



10 

inhibit the interaction by 100 percent, maximum inhibition 
was only 68 percent (35). The most extrane example in v^ich 
hapten cannot inhibit id/anti-id interactions are in those 
studies in which cross reactive ids are present on antibody 
molecules of widely different specificity. For example, 
Eichmann et al. (36) showed that one half of the A5A id 
producing clones in A/J mice immunized with a streptococcal 
carbohydrate lacked the ability to bind this antigen. 
Obviously then, antigen vrould not be expected to inhibit 
this id-anti-id interaction. In other studies. Bona et al. 
(37) showed that not all the id positive antibody following 
i'Timunization with inulin could be ranoved with an inulin 
immunoabsorbent . In these experiitients , the anti-inulin anti- 
body produced followiing antigenic stimulation bears a 
predominant id. However, sane imnunoglobulin following 
antigenic stimulation had this id but lacked specificity for 
inulin. These experiments tlierefore suggest that scire 
nechanism exists naturally in vdnich id positive clones of 
immunoglobulin producing cells are expanded following 
antigen stimulation but that not all the id positive immuno- 
globulin is specific for the itimunizing antigen. These 
experiments clearly show tliat although id/anti-id can 
usually be hapten inhibited, this property is not a require- 
ment for an antibody to be ant i- idiotypic. 



11 

Identical ids have been found irrespective of the 
isotype of the antibody. The mechanism by which IgE and IgG 
antibody can have identical idiotypes relates to tlie gene 
rearrangement that occurs during differential expression of 
heavy chain genes ( 38 ) . As a single clone of cells goes 
through isotypic shift, a single variable region of the 
genes v*iich includes the idiotype, will become linked to 
various heavy chain gene fragments. A single cell will 
differentiate into plasma cells which express different 
heavy chain genes but the same variable gene sequence ( 39 ) . 
Therefore, it is possible for ids to be shared between anti- 
bodies of the satre binding ability irrespective of the 
isotype. This implies that regulation of IgG antibody by 
anti-id networks may also result in IgE antibody regulation. 

Idiotypic determinants are usually defined sero- 
logically. There are a number of different ways to produce 
anti-id (reviewed in 40-42). Anti-id can be produced across 
tile species barrier, within the same species, witliin the 
same strain, or more importantly, even within tlie !^ne 
individual that produced the id. Ant i- id have been used to 
determine if the id of the antibody molecule may have a 
function otlier tlian to bind antigen. This has been done by 
examining v^at functional significance the presence of 
ant i- id had on the corresponding id in vivo . 



12 

There are nuirerous reports that have shown that the 
passive administration of anti-id or the active induction of 
anti-id results in the suppression of tlie corresponding id 
(reviewed in 40-46). This ixtxiulation acts directly on 
B-cell3 or indirectly through T-cells. For example, in a 
B-cell tumor model, Balb/c mice irrmunized with MOPC 315 
iTYeloma protein produced antibody with specificity for the 
id of MOPC 315. Subsequently these mice were injected with 
a jyiOCP 315 bearing plasrtHcytoma and the tumor growth was 
inhibited. It has also been shown tiiat the immunization of 
i^PC 315 protein also induces idiotype specific T-suppressor 
cells that inhibit the MDPC 315 tumors secretion in vivo 
(47). Cosenza and-Kohler (48) demonstrated that anti-id can 
act as an anti-antigen receptor antibody and specif icially 
inhibit the induction of a priiiary immune response. In 
otiier studies by this same group, anti-id, vAiich was 
specific for anti-phosyphorylcholine (PC) antibody, signif- 
icantly inhibited anti-PC plaque forming cells to a degree 
similar to the inhibition seen with antigen ( 49 ) . 

These studies show tliat experimentally, the admin- 
istration of anti-id or immunization with id to induce 
anti-id can result in id suppression. However, if anti-id 
regulates id during a normal immune response, auto-anti-id 
should be part of the response. 



13 

A number of studies have shown tiie presence of auto- 
anti-id during a nomnal immune response to an antigen 
(50-57). Bankert and Pressman (50) showed that an antibody 
with auto-anti-id activity could be detected in rabbits 
during primary and secondary immune response to both sheep 
red blood cells and to the hapten, 3-iodo-4-hydroxy-5- 
nitrophenyl-acetic acid. Kelsoe and Cerny (51) have demon- 
strated a reciprocal expansion of antigen activated idiotype 
bearing clones of lymphocytes followed by expansion of 
clones which bear anti-id receptors in Balb/c mice immunized 
to Streptococcus pneumonia . They hypothesized that the 
out of phase expansion of the reciprocal cell sets was the 
result of interactions of id and anti-id. The production of 
auto-anti-id in wan has been demonstrated to occur during 
the ifiTOune response against tetanus toxoid. The presence of 
this anti-id was associated with the loss of some of the 
anti-tetanus toxoid idiotypes ( 52 ) . Naturally occurring 
anti-id has also been demonstrated in myasthenia gravis 
patients using, as tlie idiotype probe, a mouse monoclonal 
antibody. Those patients with the highest titer of anti- 
receptor antibody had the lowest level of anti-id, while in 
patients with the lowest titer of anti-receptor antibody 
(id), the highest titer of anti-id was detected (53). 
Comparable findings have been reported in patients with 
anti-DNA. antibody and reciprocal anti-id in systemic lipus 



14 

erythonatosus (54) and in sane IgA-def icient people in terms 
of anti-casein antibody and its reciprocal anti-id ( 55 ) . 
In these later experiments the ant i- id was detected using 
hannologous antibody as the id probe. 

These experiments suggest that because anti-id is 
present during a normal immune response and regulates the 
expression of ids, anti-id nay he an important part of the 
regulation of the immune response. 

In reference to IgE, Geczy and his associates (58) have 
shown that in guinea pigs, the administration of synge- 
neically derived antibody led to a marked suppression in tiie 
IgE level as measured by passive cutaneous anaphylaxis. 
This treatment also resulted in the production of anti-id 
and if this anti-id was given to a guinea pig followed by 
antigen stimulation, tliere was a rrarked suppression in tiie 
subsequent response. This group has shown that in the 
mouse, the preexisting anti-hapten IgE and IgG antibody 
response could be suppressed with either anti-hapten or 
anti-carrier ant i- idiotypic antibody (59-61). 

These experiments and others like tliem show that 
id/anti-id interaction results usually in suppression of the 
immune response. However, this is not always the case. For 
exaiTiple, Eichmann and RajewsVzy (62) showed that the ■■ • 
injection of guinea pig IgG-j^ anti-id would enhance the 
expression of id designated ASA when stimulated with 



15 

Streptococcus viiereas if the anti-id was an 19^2' 
expression of ASA id -was suppressed. Recently, Forni et al. 
(63) showed that the injection of anti-SRBC IgM into nornnal 
mice induced plaque forming cells of the same specificity as 
the injected antibody. Furtlier analysis established that the 
mechanism of this enhanced responsiveness was based on 
id/anti-id interactions (53,64). The authors state that 
"these results support network concepts. Thus if an antigen 
specific response can be induced solely by using components 
of the immune system itself, it follows that, in its basic 
econon^, this system is autonomous and does not depend on 
the introduction of antigen to adjust to new dynamic states" 
(63 p. 1127). In tliis case anti-id most probably acted as 
an internal mage of antigen. There have been other 
examples that demonstrated the mimicry of antigen by anti- 
body. For example, Sege and Peterson (65) showed that 
anti-id prepared against antibody to insulin could mime tiie 
action of insulin in cells. Schreiber et al. (66) showed 
that anti-id against rabbit antibodies to alprenolol would 
oonpete with alprenolol for tlie binding site on turkey red 
blood cells. This anti-id could also stimulate adenylate 
cyclase activity in the cells. 

This discussion raises the possibility that the admin- 
istration of autologous antibody might regulate antigen 
specific IgE response in the dog through id/anti-id 



16 

networks. Therefore the objectives of the work presented 
here were 

1) To develop a consistent IgE antibody response in 
the dog and to study the kinetics of this response. 

2) To examine the effects that autologous antibody 
administration had on an ongoing IgE response. 

3) To determine if an anti-id response occured at any 
point during the experiment and if so, to examine the 
relationship between ids and anti-ids. 



CHAPTER 'I\VD 
1HE INDUCTION AND KINETICS 
OF AN ANTI-DNP IGE RESPONSE 



Introduction 

The value of the dog as an experimental model to 
study atopy has been described. However, the expense and 
difficulty of obtaining atopic dogs necessitated the 
development of a system in v^ich antigen-specific IgE could 
be consistently induced. The use of a hapten-coupled 
carrier as an antigen vas felt to be more convenient than a 
more complex, heterogenoias substance such as an allergen to 
study the synthesis and regulation of IgE antibody. 
Furthermore, Halliwell (7) and Schwartzman et al. (67) have 
shown that two dogs immunized with dinitrophenol coupled to 
ascaris antigen and administered in aluminum hydroxide as 
tlie adjuvant, developed anti-DNP IgE antibody. However, it 
is not known a) if all dogs so immunized produce IgE anti- 
body, b) how long the detectable IgE response remains, and 
c) what the immune response in terms of other isotypes 
might be. The purpose of tlie following experiments, then, 
was to induce a consistent anti-hapten IgE antibody 

17 



18 

response and to examine the kinetics of the IgE, IgG and 
IgM anti-hapten antibody response. 

Materials and Methods 

Protein Concentration Determination 

The concentration of immunoglobulin was determined 
from known molar extinction coefficients and by its ability 
to absorb light at 280 nm. Alternatively, the protein 
concentration was determined at 595 nm using Bradford's 
reagent (68) and interpolated from a standard curve derived 
from the absorption values of a series of dilutions of a 
similar freeze dried purified protein of known concen- 
tration. The measuranents with both techniques gave concor- 
dant results. 

Antigens 

Azobenzenarsonate coupled to keyhole limpet hemo- 
cyanin (ABA-KLH) was a gift from Dr. Mark Greene, Harvard 
University. Ascaris antigen was prepared from adult 
Toxocara canis by the method of Strejan and Campbell 
(69) and modified as follows: Fifty adult canis 
were obtained from the gastrointestinal tract of euthanized 
dogs. The worms were washed with fiiosphate buffered saline 
(PBS), pH 7.2, containing 0.02 percent sodium azide, ground 
with a mortar and pestle and incubated for 48 hours at 4° 
C. Large particulate matter was removed by centrifugation 



19 



at 1000 X g for ten minutes in an lEC centra-7R centrifuge 
(International Equipment Co.) The supernatant was then 
centrifuged at 49000 x g for one hour in an L8-70 ultra- 
centrifuge (Beckman Instrument Co., Norcross, C^.) to 
ranove fine particles and was then chronatographed through 
a Sephadex G-lOO column (Pharmacia Fine Chemicals, 
Piscataway, N.J. ) . The first peak was pooled, concentrated 
by negative pressure dialysis, dialyzed against PBS, pH 
7.2, passed through a filter having 0.2 micron pores filter 
(Acrodisc, Gelman Co., Ann Arbor, Mi.) and used as the 
ascaris antigen (ASC). Humn serum albumin (HSA) fraction 
V was obtained from Sigma Chonical Co. (St. Louis, Mo.). 
Bovine gamrra globulin (BGG) was prepared from serum of an 
adult cow by precipitation with 40 percent saturated 
ammonium sulfate. The precipitate was dialyzed against 
0.035 M phosf^ate buffer, pH 8.0 and was then chroma to- 
graphed through a diethylaminoethyl cellulose (DEAE) ion 
exchange column (DEA, DE52, Whatman Chemicals, Kent, 
England) equilibrated with this same buffer. The effluent 
protein was concentrated by negative pressure dialysis and 
dialyzed against PBS, pH 7.2. 

Dinitrophenylation of Proteins 

Dinitrophenylation of protein was performed by mixing 
equal weights of protein, potassium carbonate (Fisher 
Scientific Co., St. Louis, Mo.) and 



20 



2,4-dinitrobenzenesulphonic acid (DNP) (Easta^n Kcxiak Co., 
Rochester, N.Y.) were mixed in distilled water (70). This 
was then incubated while gently stirring for 18 hours at 
room temperature. The solution vas chrotiatographed through 
a Sephadex G-25 column (Pharmacia Fine Qiemicals, 
Piscataway, N.J. ) to separate bound fran free DNP, The 
dinitrophenylated protein was concentrated by negative 
pressure dialysis and extensively dialyzed against PBS, pH 
7.2. The extent of substitution vvas estiiiated by neasuring 
light adsorption at 360 nm and assuming a molar extinction 
coefficient of 1.75 x 10 for the dinitrophenyl group. 
The average epitope density expressed as molecules of DNP 
per molecule carrier was DNP^^-HSA, DNP^^ g-BGG. Since 
A3C extract \^as a complex mixture of proteins, the extent 
of substitution was expressed as moles DNP/mg ASC and was 
6.32 X 10"^ DNP/ASC. A single batch of each of these 
antigens *as prepared and used throughout tlie experiment. 
These antigens, when not in use, were stored at -70°C. 
The degree of substitution did not change due to storage. 

Aluminum hydroxide Precipitation of Protein 

Aluminum hydroxide precipitation of protein was 
performed by mixing one part of a 5 percent sterile 
solution of aluminum potassium sulfate (AlK (SO^)^), 
iMallincrodt, Paris, Kentucky) with five parts of 1 mg/ml 



ZL 



solution of protein (70). The pH was then adjusted with 0.1 
N NaOH to pH 6.3 to ensure adequate precipitate. 

Affinity Chrcanatography 

Sejtiarose 4B beads (Pharmacia Fine Chonicals) were 
activated using cyanogen bronide (CnBr) by adding 1.5 grams 
CnBr in 20 ml distilled water to 10 ml of washed Sepharose 
4B beads and adjusted to pH 11 with I N NaOH. This mixture 
was mintained on ice at pH 11 for 6 minutes after ^ich 
the beads vere washed with 100 volumes of iced cold water. 
Ninety milligrams of protein in 6 ml PBS, pH 7.2 were added 
and incubated for 12 hours at 4°C. Alternatively pre- 
activated Sepharose 4B beads were obtained (Pharnacia Fine 
Chanicals) and protein vas bound to these beads as 
described by tlie nanuf acturer . To ranove unbound protein 
in both cases, the beads were washed with five alternate 
cycles of 0.1 M Tris buffer, pH 8.3 containing 0.5 M NaCl 
followed by 0.1 M glycine HCl, pH 2.8. Any ramaining sites 
were blocked by incubating tlie beads in 0.1 M Tris buffer, 
pH 8.3 for four hours at roan temperature. The column vss 
then flushed with normal canine serum and washed as 
described above. 



22 



Pepsin Digestion and Purification of F(ab)' ^ Antibody 
Fragments . — 

The usual procedure for F(ab)'2 digestion of immuno- 
globulin vas to digest the antibody with 6 percent pepsin 
(w/v) in 0.2 M acetate buffer, pH 4.5 for 18 hours at 37 
°C. However, this process resulted in some loss of 
antigen binding of the F(ab)'2 presumbly fran the 
prolonged incubation time at pH 4.5. Where maintenance of 
this activity was critical, protein was digested with 20 
percent pepsin w/v in 0.2 M acetate buffer pH 4.5 for five 
hours at 37°C. The digested protein was separated from 
Fc pieces and intact antibody by passage through a cyanogen 
bromide-activated heavy chain specific immunoabsorbent 
column followed by passage through a Staphylococcus 
protein A affinity column (Pharmacia Fine Chemicals). The 
effluent was concentrated by negative pressure dialysis and 
dialyzed against PBS, pH 7.2. 

Antisera 

a) Preparation and purification of anti-IgG. Normal 
canine serum (^iCS) was precipitated with a 40 percent 
saturated solution of amiiK)nium sulfate. The precipitate 
was dialyzed against 0.035 M phosphate buffer, pH 8.0 and 
applied to a DEAE ion exchange column equilibrated with 
this same buffer. The effluent protein was concentrated by 



negative pressure dialysis. One milligram of this material 
was emulsified in complete Freund's adjuvant (CFA) and 
administered intramuscularly to ral±)its at two week 
intervals four times. Fifty milliliters of blood were 
obtained from the rabbit by ear vein venapuncture every two 
weeks starting after tlie second immunization. All serum 
which gave visible precipitation reactions by agar-gel 
diffusion against canine IgG was pooled. This antisera 
vas passed through a cyanogen bromide-activated sepharose 
4B F(ab)'2 affinity column, to renove light chain 
activity, followed by adsorption to and elution with 
alycine HCl (Osl M) , pH 2.8 frcm a canine IgG bound 
affinity column. This anti-IgG detected three subclasses 
of canine IgG (IgG^, IgG2^j^, 15^2^) but no other 
protein as measured in an immunoelectrophoresis (70) of NCS 
(figure 1). To determine if this antiserum detected IgE, 
the antiserum was radiolabelled and used in a radio- 
immunoassay. The serum sample tested contained both anti- 
DNP IgG and anti-DNP IgE. Therefore, an aliquot of this 
serum vas heat inactivated and the level of anti-DNP IgG 
was compared in this aliquot to a second aliquot of this 
serum tliat vias not heat inactivated. Additionally, anti- 
canine IgE was added to an aliquot of this sample to 
determine if this unlabelled anti-IgE might ccmpete with 
the anti-IgG for Fc binding sites. Heating serum for four 
'nours at 56°C destroys the heavy chain antigenic 



determinants of canine IgE (7). The level of anti-DNP 
antibody increased both when the semm vas inactivated and 
vdnen non- labelled anti-IgE antiserum was added to the 
sample. This indicates that this anti-IgG antiserum has 
minimal, if any, anti-IgE activity. 

b) Preparation and purification of anti-IgE. A 40 per- 
cent saturated ammonium sulfate precipitate of serum 
obtained, from a dog that was heavily parasitized and 
presumed to have high levels of IgE, was dialyzed against 
0.035 M ^osphate buffer pH 8.0 and applied to a DEAE 
cellulose column equilibrated with this same buffer. The 
effluent protein was concentrated by negative pressure 
dialysis and applied to a set of three in series Sephacryl 
S-200 columns (Pharmacia Fine Chemicals, Piscataway, N. J. ) . 
The first one-third of the second protein peak, which was 
the IgE-rich fraction as determined by agar-gel immuno- 
precipitation, was collected, concentrated by pressure 
dialysis and reapplied to these columns. The resulting 
IgE-rich fraction was collected and used to immunize 
rabbits as described previously. The rabbits were bled as 
described above. Serum that produced visible precipitation 
lines against the immunizing antigen in an agar gel immuno- 
diffusion were pooled. The resulting antiserum detected 
both IgE and IgG by immunoelectrophoresis . It was rendered 
specific for tlie former protein by passage tlirough an 



Figure 1. 



The specificity of anti-canine IgG as assayed in an 
iinmunoelectorphoresis against normal canine serum. 



Figure 2. 

The specificity of anti-canine IgE as assayed in an 
iinmunoelectrophoresis against normal canine serum 
(bottom well) and this same serum after heat 
inactivation (top well). 



Figure 3. 

The specificity of anti-canine IgM as assayed in an 
Immunoelectrophoresis against normal canine serum. 
The anti-canine IgM in the bottom through is before 
adsorption with the supernatant of a 50 percent 
satiorated amcroniutn sulfate precipitate of normal 
canine serum. The top trough has the anti-canine IgM 
antiserum after this treatment. 



25 




27 

affinity colunm made with the heat inactivated iitiiTiunogen 
which removed all antibody except anti-IgE antibody. 
Purified antibody was then prepared by adsorption to and 
elution from an IgE-rich affinity column. This purified 
antiserum detected a single heat-labile protein by inimuno- 
electrophoresis (figure 2), produced reverse cutaneous 
anaf^ylaxis in dogs at a high dilution of serum (10~^) 
and was unable to detect canine anti-DNP IgG in a RIA 
indicating that it had no specificity for tliis antibody. 

c) Preparation and purification of anti-IgM. Canine IgM 
myelona serim, ^vhich contained approxirrately 58 mg/ral IgM 
was chraTBtograc±ied on Sephacryl S-200 and the void volurre 
was collected to obtain IgM. Two milligrams of this 
material was emulsified in CFA and injected intra- 
muscularly at four sites into sheep. This was repeated at 
two week intervals five times. Five hundred milliliters of 
blood were collected by jugular vein venapuncture every two 
weeks. Sera that produced precipitation lines against the 
immunizing antigen, in an agar-gel diffusion against the 
immunizing antigen, were pooled. Light chain activity vas 
rsTVoved from tlie antiserum by passage tlirough a canine IgG 
affinity column. Antibody was then purified by adsorption 
to and elution from an IgM affinity column. The eluted 
proteins produced two bands on immunoelectrophoresis of 
ISCS, one of which was IgM and the other an unknown protein. 



28 

Itiis second activity v^as removed by adsorption with the 
supernatant of a 50 percent saturated anmonium sulfate 
precipitation of m:S (figure 3). This antiserum vvas 
assayed for anti-DNP IgE and IgG activity by RIA. Serum 
that was used contained both of these antibody isotypes. 
No antibody ^^as detected indicating the antiserum did not 
have activity for IgE or IgG. 

Isotope Labelling of Protein 

Two methods were used to label proteins with radio- 
active iodine. In the first mstiiod, between 1 and 2 mg of 

protein in 0.1 ml PBS pH 7.0 without azide and .5 mCi 
125 

I (Amersham, Chicago, II.) was incubated on ice with 
15 \il KI(0.1 m) and 30 ^1 chloramine T (10 nM) for 15 
minutes. After tliis incubation, 25 ^il sodium netabi- 
sulphate (10 mM) and 50 |il XI (100 mM) were added to stop 
tlie reaction. To separate bound and free iodine, the irat- 
erial vas chroma tographed through a G-25 Sephadex column. 
The first peak containing radiolabel was pooled, concen- 
trated and dialysed against PBS, pH 7.2. Alternatively, 
one lodobead^ (Pierce Chemical Co., Rockford, II.) was 

added to 100 ^g of protein in PBS, pH 7.0 and 0.5 mCi 
125 

I. After a fifteen minute incubation, the bound and 
125 

unbound I vas separated as described earlier. The 
specific activity of the radiolabelled antibody was usually 
about 300 |jCi/mg protein (range 212-496). 



29 



Radioiinmunoassay for the Detection of DNP Specific 
Antibody 

Microtiter wells (ImniiLon Ranov-a-well Strips^, 
Dynateck, Richmond, Va.) were coated with 50 ^il of 20 
Hg/ml dinitrophenylated bovine garmB globulin (DNP-BGG) in 
Tris buffer, pH 8 (0.1 M Tris, 0.15 M NaCl). After incu- 
bating for 12 hours at 4°C the wells were then washed 
three times with this buffer. Any remaining sites were 
blocked with 2.0 percent HSA in PBS containing 0.5 percent 
Tween 20 for three hours at room temperature. Phosphate 
buffered saline, pH 7.2 containing 0.5 percent Tween 20 and 
2.0 percent HSA is referred to as RAST+ and this same 
buffer without HSA is called RAST-. Serum samples were 
diluted in PBS, pH 7.2 added to the appropriate wells, 
incubated for tliree hours at 4°C followed by five washes 
with RAST-. Approximately 50,000 counts per minute (cpm) 
of radiolabelled antiserum in RAST- was added to the well, 
incubated for three hours at and washed tliree times 
with RAST-. The radioactivity associated with each well 
was determined in a Searle- Packard gamma counter (Chicago, 
II.). Each sample was assayed in triplicate and each 
sample was counted for one minute. The maximum numter of 
cpm bound was about 20% of tlie anount added. The back- 
ground activity was determined by including in each assay 
the following controls: 1) A set of triplicate wells in 
which BGG rather than DNP-BGG was used as the antigen, 2) A 



30 

triplicate set of wells in which PBS rather than serum was 
added. The mean cpm fron these controls were subtracted 
from the cpm of the test sample. Although the values 
varied fron experiment to experiment, the iTHximum cpm of 
these controls were consistently lower than the lowest 
values obtained for test samples. 

A standard serum sample was included with each 
assay as an internal reference. An arbitrary antibody 
concentration was determined by assigning a value of 64 
units to tile undiluted standard IgG and IgE sample and 32 
units to the undilute IgM standard. By interpolating from 
the linear portion of the standard curve, the relative 
units of antibody for test samples were calculated. 

Animals and Imnunization Schedule 

Outbred pregnant female dogs were obtained from the 
Division of Animal Resources, University of Florida. Serum 
fran these dogs was screened by RIA to ensure that they did 
not have anti-DNP antibody at the time of whelping. 
The puppies of these bitches were used as experimental 
animals. Seram samples were obtained on the day of birth 
and weekly therafter. Each puppy received 100 (ag alaminum 
hydroxide precipitated dinitrophenol coupled ascaris 
antigen by the intraperitoneal route on the day of birth 
and at two week intervals on three further occasions. Each 



31 

dog received a distanper-hepatitis modified live virus 
vaccination at week four and eight. 



Results 

Standard Curve 

The relative antigen-specific antibody concentration 
was determined by interpolation from the linear portion of 
the standard curve included with each assay. This serum 
sample contained high levels of the isotype under inves- 
tigation. An exartple of a standard curve for anti-DNP IgE, 
IgG and IgM is given in figures 4, 5 and 6. 

Antibody Response 

Twenty-eight dogs immunized with 100 jjg D^]P-ASC in 
aluminum hydroxide developed an IgE, IgG and IgM sarum anti- 
body response. The mean relative antibody concentration 
for the three isotypes is depicted in figure 7, 8 and 9. 
The IgM response usually vas highest in samples taken seven 
days after the first injection of antigen. However, as 
seen in table 1, eight of the dogs (5,7,8,12,14,16,23,25) 
had anti-DNP IgM concentrations that were greatest in 
samples obtained at two weeks and three dogs (9,17,22) 
after tiiree weeks. Four weeks after the first antigenic 
challenge, five dogs had no detectable IgM antibody and 



33 




LO O LO O LO O LO 

ro ro c\j c\J — — 



001 ^ ^do 



a; 



35 



o 

00 
CM 




CO CD (\J 

0001 X ^do 



Figure 6. 



Dilutions of the standard anti-DNP IgM serum sample 
assayed by RIA. The bars represent the standard 
deviation of the mean. 



37 




5 10 20 40 80 160 320 
Reciprocal of Dilution 



38 

after six weeks the levels of IgM antibody fell to back- 
ground despite naintenance of the immunizing protocol. 

The anti-Dl^ IgE and IgG antibody responses followed 
similar kinetics to each other. There an initial lag 
of two weeks before antibody of these classes was detected 
(figures 8 and 9). At the time of the second immunization 
(two weeks after the primary immunization), there was a 
sharp rise in the antibody levels v^^iich continued for one 
additional week. Thereafter, the antibody concentration 
vas maintained at that level or started to gradually 
decline. As vas the case in the IgM antibody response, 
seme dogs deviated from tlie general trend. Two dogs had 
detectable IgE antibody levels one week after primary 
immunization (Table 2) wiereas three cisgs failed to 
develop a detectable response until after the third week 
and, in the case of one dog, IgE antibody was not detected 
until the fifth week frcxn primary immunization. The IgE 
antibody response persisted through the seven week course 
of tlie experiment in all dogs. Anti-DNP IgG antibody vvas 
detected in nine dogs one week after primary immunization 
(table 3) and by the fourth week, all dogs had an IgG anti- 
body response. Detectable IgG persisted throughout the 
immunization schedule but there was a gradual decline in 
IgG antibody levels towards tlie end of the iiimunizing 
schedule (See figure 9, table 3). 



a; 



■H rH 
4-> -H 




4-> 4-1 H 

^ (C „ 

4-) J-1 Z 

C C Q 

0) 0) I 

O -H 

S S 

8 8 § 



2 oofsooooo 



ro ro O) o — 
cn ^ ro ^ • 

-H 4^ -H -H -H 




CM O 00 CD 'sT CO 

uoiiDJjUQOuoQ Apoqijuy aAjioiey 



4-) 
4J 



(0 > 0) 

M r-| U 



8 



4J cn 



0) Cn<fi >, 

•H CU (0 

(C i= Q +J 
O 



4J 

to OJ 




4 2 



^ GO C\J 

b 00 



CM 



o> ^ N 
p I I I I I 

«b o o o o o 



I 1 I 

ro CD 00 



CS 0) O 0^ CD ^ 

(X cO id ro lO ro lO 

-H44-H+1+I-H-H 

c o— CD — cMcnror^ 

O Osl C£) id CD ^ CD 

^ O — CVJrO^lOCDN 



CX) 




OJ O 00 CD ^ (\J 

silun Apoqiiuv aA!iD|9y 



O 0) 



CT'. 
0) 
3 



4J 



C 



5 



< o 

M 

CD 

^ Si 



Relative Antibody Units 



r\34^(j)OOOf\34^CD 



CD 
CD 

CO 




45 



Table 1 

The Relative Anti-DNP IgM Concentration in 28 Dogs 



Animl 

Number 1 2 3 4 

Weeks " 



1 16.6+.87 4.4+.39 6.5+. 47 4.5+. 06 

2 10.6+1.23 3.2+. 23 3.4+. 24 4.4+. 27 

3 .3+0 1.4+.56 .6+. 14 

4 .2+. 01 .2+0 .1+.03 

5 

6 

7 
Aniiral 

Number ^ §. 7 8 

Weeks ~ ~ 



1 2.9+.25 6.4+.82 6.8+1.01 3.1+.28 

2 5.8+.90 5.6+. 39 7.8+.32 5.2+.17 

3 5.1+.07 3.4+. 31 

4 3.9+. 24 

5 

6 

7 

Aniitval 

Number 9 10 11 12 
Weeks 



1 10.9+1.03 7.0+.02 5.0+.25 

2 2.0+.14 3.0+.41 2.8+.46 5.1+.37 

3 3.3+.29 .8+.01 1.8+.19 2.0+.21 

4 1.0+.07 1.8+.27 

5 ~0 

6 

7 

AniiTHl 

Number 13 14 15 16 
Weeks 



1 5.5+.16 1.8+.18 10.1+.79 1.3+.12 

2 5.0+.23 5.2+.33 6.1+.43 2.6+. 37 

3 4.8+. 07 2.3+. 16 1.8+109 2.5+. 14 

4 .9+.13 .5+. 22 1.6+.17 1.2+.20 

5 

6 

7 



46 



Table 1 Continued 



Animal 










Numter 


17 


18 






Weeks 






















n 


1 


1.1+.03 


14.8+2.01 


3 9+ 39 




2 


4.4+. 17 


5 . 6+ . 81 


2 1+ 09 


^ • J. 1 . *4U 


3 


6.0+.31 


3.1+.23 


• Or a UZ, 


n 


4 


1 . 7+ . 15 


2 1+ 10 


u 


u 


5 


.8+.0 


~o 





n 


6 


~o 








n 

V 


7 











fi 

U 


Animal 










Nuinber 


21 


22 


Co 


0/1 


Weeks 






















n 


1 


8.7+. 83 


.7+. 15 


8.1+.44 


5 1+ 46 


2 


7.9+. 09 


4.5+. 18 


12 6+ 99 




3 


3.2+. 48 


6.2+. 46 


2 7+ 23 




4 


1.6+.06 


2.1+.05 


~0 


n 


5 


.1+0 


~o 





n 


6 











n 

u 


7 














Animal 










Number 


25 


26 


27 


28 


Weeks 

























1 


4.0+.09 


13. 8+. 1.35 


5.2+. 29 


10.6+.87 


2 


6.2+. 21 


9.1+1.37 


.7+0 


X 


3 


2.5+. 30 


3.4+. 40 


1.0+.16 


4.2+. 36 


4 





2.1+.11 





.8+. 023 


5 














6 














7 















a) Each dog was inrounized with DNP-ASC in adjuvant at weeks 0,2,4 
and 6. 

b) The units were calculated from a relative antibody 
concentration scale derived from the titration of a serum sample 
containing anti-DNP IgM. A value of zero indicates no detectable 
anti-DNP IgG. The data was the n^n antibody concentration of a 
sample run in triplicate + the standard deviation from the mean. 
This was calculated by adding and subtracting the standard 
deviation to the n^n cpn and calculating tlie relative antibody 
concentration for these numbers. These numters were tiien 
subtracted from the mean concentration. 



47 



Table 2 

The Relative Anti-DNP IgE Concentration In 28 Dogs 
Following Iirmunization with DNP-ASC a) 



Animal 
Number 
Weeks 



1 

2 

3 

4 

5 

6 

7 

Aninnal 
Number 
Weeks 



1 

2 

3 

4 

5 

6 

7 

Animal 



b) 






1.2+.31 
1.5+.27 
1.7+.16 
1.3+.49 
1.0+.22 
3.0+.96 





.6+. 03 
3.6+. 46 
11.9+.99 
7.8+1.04 
9.0+.63 
11.0+.71 
14.1+1.42 






10.8+1.90 
11.3+.27 
9.2+. 36 
11.6+.24 
5.6+. 70 
7.7+1.01 

6 




1.9+.17 
13.4+1.35 

8.0+.61 
12.7+.62 
16.1+.90 
12.8+.88 







3.2+. 36 
10.0+.76 

8.5+. 44 
10.1+.83 

3.3+. 12 

7.0+.41 






3.2+. 10 
11.8+.77 

6.7+. 83 
11.8+2.45 

5.5+. 69 

2,8+. 25 







3.2+. 21 
4.8+. 07 
4.6+. 29 
5.1+.36 
3.3+. 16 
4.9+. 37 






,6+. 07 
,0+1.00 
.9+. 43 
15.2+.18 
7.6+. 45 
6.6+. 51 



5. 
3. 



Number 


9 


10 


11 


12 


Weeks 























1 


1.7+.39 











2 


5.6+. 26 





2.1+.24 


7.5+. 88 


3 


18.2+2.73 


3.1+.69 


8.3+. 74 


13.2+1.86 


4 


8,4+. 97 


5.5+. 43 


7.8+. 79 


9.1+1.02 


5 


10.0+.64 


5.2+. 47 


3.9+. 81 


26.8+1.30 


6 


7.1+.31 


3.0+.30 


2.4+. 65 


7.0+1.51 


7 


8.7+. 51 


12.6+.76 


4.8+2.38 


15.4+.36 


Animal 










Numter 


13 


14 


15 


16 


Weeks 

























1 














2 





1.8+.36 


l.O+.ll 





3 





9.5+. 37 


3.4+. 48 


.7+. 12 


4 





8,4+. 67 


7.5+1.06 


3.9+. 61 


5 


.3+. 11 


12.0+.87 


5.1+.56 


3.9+. 14 


6 


.8+. 23 


8.5+. 21 


.5+. 01 


3.7+. 37 


7 


1.6+.52 


19.3+1.98 


5.2+. 50 


4.5+. 66 













48 



Table 2 Continued 



Animal 










Number 


17 


18 


19 


20 


Weeks 

























1 














2 


.2+. 07 


10.1+.21 





3.3+. 09 


3 


11.0+.96 


9.3+. 49 


1.4+.08 


4.2+. 15 


4 


7.7+. 14 


8.9+. 47 


.3+. 04 


4.8+. 33 


5 


9.8+. 34 


9.4+. 67 


1.0+.20 


5.2+. 40 


6 


2 . 2+. 26 


6.9+. 52 


1.5+.08 


5.8+. 26 


7 


6.7+. 31 


15.3+.89 


2.3+. 21 


7.1+,43 


Animal 










Number 


21 


22 


23 


24 


Weeks 

























1 














2 


3.0+.10 


1.5+.17 


2.1+.13 


.1+.04 


3 


4.5+. 11 


2.6+. 12 


1.7+.09 


1.0+.06 


4 


4.G+.27 


3.0+.31 


1.8+.14 


1.0+.17 


5 


5.6+. 36 


2.6+. 21 


1.0+.13 


1.6+.31 


6 


2.3+. 12 


2.8+. 32 


0.8+.07 


0.9+.07 


7 


4.8+. 22 


2.6+. 60 


0.9+.21 


1.3+.29 


Am ma 1 










Number 


25 


26 


27 


28 


Weeks 

























1 














2 


1.0+.29 


1.7+.37 


.6+. 05 


X 


3 


2.7+. 31 


2.1+.05 


3.2+. 22 


1.1+.06 


4 


2.0+.26 


3.0+.42 


2.0+.23 


7.3+. 26 


5 


1.8+.19 


1.3+.12 


1.0+.15 


8.8+. 51 


6 


.7+. 09 


4.3+. 27 


1.5+.13 


3.3+. 16 


7 


1.2+.10 


1.7+.28 


0.8+.06 


8.4+. 42 



a) Each ±)g ms immunized with DNP-ASC in adjuvant at weeks 0,2,4 and 6. 

b) The units were calculated from relative antibody concentration scale 
derived from tlie titration of a serum sample containing anti-DNP IgE. A 
value of zero indicates no detectable anti-DNP IgE. The data was the 
mean antibody ooncentration of a sample run in triplicate + the 
standard deviation from, the mean. This was calculated by adding and 
subtracting the standard deviation to the mean and calculating the 
relative antibody concentration for this number. The relative 
concentration for this number was subtracted fron the mean concentration. 



49 

As TOuld be expected for out bred animals, there was 
considerable variation in the immune response between dogs. 
If however, the IgG antibody concentration of dogs within 
single litters are examined, a more homogeneous trend is 
observed (table 4). There was, however, considerable 
aniraal-to-aniiTQl variation within a litter in the level of 
antigen specific IgE and IgM (tables 5 and 6). If these 
two litters are compared statistically, at each thae point, 
using a student T test, there is a significant difference 
in the mean antibody level between the two groups in the 
IgG antibody after the first week (P is less than 0.001 in 
all instances ) . 

When all the animals are considered, it appears that 
some are generally high responders to the antigenic stimu- 
lation whereas the response in other dogs is low. The high 
response or low response is seen for both IgG and IgE 
antibody classes in a single animal. For example, dogs 2 
and 6 have a strong IgE and IgG response v^ereas dogs 15 
and 24 have very weak responses. Although tliis trend 
predominates, this association of high responses or low 
responses is not always consistent, and a regression 
analysis comparing the level of anti-DNP IgG to the 
anti-DNP IgE failed to show a statistically significant 
correlation (p greater than 0.05). 



50 



Table 3 

The Relative anti-DNP IgG Concentration in 28 Dogs 
Following Ininunization with DNP-ASC a) 



Animal 
Number 
Weeks 

1 

2 

3 

4 

5 

6 

7 

Animal 
Numter 
Weaks 

1 

2 

3 

4 

5 

6 

7 

Animal 
Numter 
Weeks 

1 

2 

3 

4 

5 

6 

7 

Animal 
Numter 
Weeks 

1 

2 

3 

4 

5 

6 

7 






4.1+.21 
3.7+. 20 
6,0+. 64 
8.1+.33 
7.9+1.05 





0.3+.02 
3.7+. 12 
8.6+. 17 
9.1+.34 
7.4+. 23 
8.6+. 61 







7.3+. 69 
15.4+1.71 
18.5+.86 
16.0+.46 

8.3+. 62 

13 





10.1+.61 
9.3+. 75 
7.8+. 26 
5.0+.16 



b) 






9.1+1.4 
15.5+.20 
16.4+1.26 
26.0+1.33 
23.3+. 14 

6 





9.1+1.21 
15.3+.86 
17,0+1.41 
25.6+. 73 
24.1+1.72 

10 





4.3+. 05 
4.3+. 12 
5.6+. 09 
7.1+.51 
6.8+. 19 

14 




5.3+. 31 
16.7+1.1 
14.3+.42 
14.7+.32 
10.4+.6 





.8+. 02 
2.4+. 21 
8.0+.10 
4.8+. 26 
8.6+. 23 
6.7+. 16 






12.2+.77 
11.1+.39 
9.2+. 64 
9.3+1.07 
5.9+. 26 

11 





8.3+. 80 
12.4+.64 
7.9+. 32 
7.5+. 93 
5.5+. 68 

15 




3.8+. 43 
8.8+. 16 
9.9+. 63 
8.9+- 91 
9.7+. 46 






3.7+. 11 
7.4+. 66 
7.4+. 14 
4.8+. 26 
6.1+.88 

8 



0.1+0 
25.4+1.31 
27.0+1.92 
30.0+3.21 
20.1+.49 
8.7+1.26 

12 




6.3+. 29 
22.5+1.10 
20.8+.93 
17.8+1.79 
10.4+.23 

16 



.3+. 02 
6.0+.11 
4.4+. 70 
11.6+1.60 
11.1+1.13 
17.2+.32 



51 



Table 3 Continued 
Animal 



Number 


17 


18 


19 


20 


Weeks 










1 














2 


2.6+. 13 


.9+. 04 


.3+0 





3 


1.7+.14 


1.7+.13 


4.8+. 17 


11.7+.11 


4 


4.8+. 36 


22.7+. 81 


16.9+.65 


12.8+.96 


5 


6.2+. 58 


24.9+1.65 


21.3+.17 


8.9+. 04 


6 


6.9+. 31 


21.8+.77 


20.8+1.16 


7.6+. 47 


7 


5.3+. 40 


28.2+1.24 


21.9+.84 


9.8+. 36 


Animal 










Number 


21 


22 


23 


24 


Weeks 










1 














2 





7.1+.33 


12.9+.92 


3.6+. 24 


3 


5.6+. 46 


7.1+.33 


12.9+.92 


3.6+. 24 


4 


5.4+. 27 


10.7+.75 


18.0+.75 


8.8+. 43 


5 


8.5+.12 


13.8+.99 


13.9+.51 


9.7+. 81 


6 


10 .6+. 93 


19 .1+1.53 


13.5+.87 


9.0+.29 


7 


7.6+. 46 


18.2+.75 


12.7+.60 


6.6+. 64 


Animal 










Number 


25 


26 


27 


28 


Weeks 








1 














2 








.8+. 07 


.4+0 


3 


1Q.9+.47 


3.3+. 33 


9.6+. 48 


X 


4 


10.0+1.79 


16.3+.84 


10.3+.68 


12.9+.85 


5 


8.2+. 97 


15.4+.93 


12.1+.06 


22.8+1.6^ 


6 


7.8+. 11 


14.7+.43 


18.7+1.78 


26.9+1.4: 


7 


5.0+.36 


10.5+.70 


19.6+.96 


28.7+1.9; 



a) Each dog received iimiunization with DNP-ASC in adjuvant at 
weeks 0,2,4 and 6. 

b) The units were calculated from a relative antibody 
concentration scale derived from the titration of a serum sample 
containing anti-DNP IgG. A value of zero indicates ro detectable 
anti-DNP IgG. The data was the mean antibody concentration of a 
sample run in triplicate + the standard deviation from the n^an. 
This was calculated by adding and subtracting the standard 
deviation to the mean cpm and calculating the relative antibody 
concentration for these numbers. This number vvas tiien subtracted 
from the mean concentration. 



52 



Table 4 

The Relative Anti-DNP IgG Concentration 
In Two Litters of Dogs 
Following Iirmunization with DNP-ASC a) 



Litter 1 





3 


4 


17 


20 


24 


25 


M +S.D b) 


Weeks 








































1 


.8 





2.6 











.6 + 1 


2 


2.4 


3.7 


1.7 


11.7 


3.6 


10.9 


5.7 + 4.3 


3 


8.0 


7.4 


4.8 


12.8 


8.8 


10.0 


8.6 + 2.7 


4 


4.8 


7.4 


6.2 


8.9 


9.7 


8.2 


7.5 + 1.8 


5 


8.6 


4.8 


6.9 


7.6 


9.0 


7.8 


7.5 + 1.5 


6 


6.7 


6.1 


5.3 


9.8 


6.6 


5.0 


6.6 + 1.7 


7 


5.4 


6.0 


3.1 


11.9 


7.2 


6.4 


6.7 + 2.9 


Litter 2 


















2 


6 


14 


19 


23 


28 


M 4- S.D. 


Weeks 








































1 











.3 





.1 





2 


9.1 


9.1 


5.3 


4.8 


12.9 


X 


8.2 + 3.3 


3 


15.5 


15,3 


16.7 


16.9 


18.0 


12.9 


15.5+ 1.8 


4 


16.4 


17.0 


14.3 


21.3 


13.9 


22.8 


17.6+ 3.7 


5 


26.0 


25.6 


14.7 


20.8 


13.5 


26.9 


21.3+ 5.9 


6 


23.3 


24.1 


10.4 


21.9 


12.7 


28.7 


20.2+ 7.1 


7 


26.8 


25.8 


18.9 


24.8 


23.0 


24.7 


24.0+ 2.8 



a) The relative antibody concentration was determined by 
extrapolation of a standard serum sample. A value of zero 
indicates no detectable antibody activity. Each dog received 
DNP/ASC in adjuvant at weeks 0,2,4 and 6. 

b) f>1ean + standard deviation 



53 



Table 5 

The Relative Anti-DNP IgE Concentration 
In 2 Litters of Dogs 
Following Inrounization with DNP-ASC a) 



Litter 1 





3 


4 


1 7 


70 






M 4- Q n V\\ 


Weeks 


































n 


n 


1 











n 







n 

U 


2 


3.7 


3.2 


2 


3 '\ 




1 
X * u 


1 Q + 1 7 


3 


10.0 


4.8 


11.0 


4 2 


1 n 


1 7 


J • O T *i . U 


4 


8.5 


4.6 


7.7 


4.8 


1.0 


2.0 


4.8 + 3.0 


5 


10.1 


5.1 


9.8 


5.2 


1.6 


1.8 


5.6 + 3.7 


6 


3.3 


3.3 


2.2 


5.8 


.9 


.7 


2.7 + 1.9 


7 


7.0 


4.9 


6.7 


7.1 


1.3 


1.2 


4.7 + 2.8 


Litter 2 


















2 


6 


14 


19 


23 


28 


M + S.D. 


Weeks 








































1 























2 


10.8 


1.9 


1.8 





2.1 


X 


3.3 + 4.3 


3 


11.3 


13.4 


9.5 


1.4 


1.7 


1.1 


6.5 + 5.6 


4 


9.2 


8.0 


8.4 


.3 


1.8 


7.3 


5.8 + 3.8 


5 


11.6 


12.7 


12.0 


1.0 


1.0 


8.8 


7.9 + 5.6 


6 


5.6 


16.1 


8.5 


1.5 


.8 


3.3 


6.0 + 5.7 


7 


7.7 


17,8 


19.3 


2.3 


.9 


8.4 


9.4 + 7.7 



a) The relative antibody concentration was determined by 
extrapolation of a standard serum sample. A value of zero 
indicates no detectable antibody activity. Each dog received 
DNP-ASC in adjuvant at week 0,2,4 and 6. 

b) Mean + Standard Deviation 



54 



Table 6 

The Relative Anti-DNP IgM Concentration 
in 2 Litters of Dogs 
Following Iimiunization with DNP-ASC a) 



Litter 1 





3 


4 


17 


20 


24 


25 


M +S.D. b) 


Weeks 








































1 


6.5 


4.5 


1.1 


2.7 


5.1 


4.0 


4.0 + 1.9 


2 


3.4 


4.4 


4.4 


2.1 


.6 


6.2 


3.5 + 2.0 


3 





.6 


6.0 








2.5 


1.5 + 2.4 


4 





.1 


1.7 











.3 + .7 


5 








.8 











.1 + .3 


6 























7 























Litter 2 


















2 


6 


14 


19 


23 


28 


M + S.D. 


Weeks 








































1 


4.4 


6.4 


1.8 


3.9 


8.1 


10.6 


5.9 + 3.2 


2 


3.2 


5.6 


5.2 


2.1 


12.6 


X 


5.7 + 4.1 


3 


1.4 


5.1 


2.3 


.8 


2.7 


4.2 


2.8 + 1.6 


4 


.2 


3.9 


.5 








.8 


.9 + 1.5 


5 

















.1 


.01 + .04 


6 























7 
























a) The relative antibody concentration was determined by- 
extrapolation of a standard serum sample. A value of zero 
indicates no detectable antibody activity. Each dog received 
inrounization with DNP-ASC in adjuvant at weeks 0,2,4 and 6. 

b) Ifean + Standard Deviation 



55 



Discussion 

The purpose of the experiments in this chapter \^as to 
induce an anti-DNP antibody response v^ich included IgE and 
to examine the kinetics of tiiis response. As described in 
the results section of this chapter, the anti-DNP IgE, IgG 
and IgM antibody response followed expected kinetics 
( 71-73 ) . The IgM response was present before IgE or IgG 
antibody was detected and disappeared after the sixth week 
in spite of continued antigenic diallenge. The IgE and IgG 
production had a two week lag period in general, but once 
they developed, they were maintained throughout the immuni- 
zation course. 

If inbred laboratory animals such as mice are 
immunized with an antigen, a homogeneous response 
ordinarily results (74). However, in species that are 
genetically heterogeneous, such as man and dogs, the immune 
response to the antigen would be expected to be nore highly 
variable ( 74 ) . In these outbred dogs there was irarked 
variation in the kinetics and magnitude of the antibody 
responses. The marked variations seen in these dogs are 
most probably the result of the genetic differences between 
them. In this context, it was notevirorthy that the IgG 
response was more homogeneous within the same litter tlian 
between litters. 



56 

The genetic makeup of the animal also influences the 
class of antibody produced following antigen challange. In 
certain inbred animal strains, it is very difficult to 
inDunt an IgE antibody response without some type of manipu- 
lative process to eliminate T-suppressor cells (75,76). 
Furthermore, if a comparison is made between allergic and 
non-allergic people, a marked difference in antigen- 
specific IgE responsiveness to certain antigens is seen. 
Those individuals with allergic tendencies will have an 
enhanced IgE response to allergens ^ereas non-allergic 
people may not develop IgE antibody (77). There were 
notable differences between individual dogs in terms of 
their IgE response and in contrast to the IgG response 
there v^as no consistent pattern within litters. It is not 
clear if this failure to see similar patterns within a 
litter in the IgE level reflects the antigen chosen to 
study IgE in these dogs or if there are multiple genes that 
govern IgE levels in c3ogs. having such a small sample 
size, a consistent pattern might not be observed for IgE 
levels. One of the dogs failed initially to develop an IgE 
titer. The IgE antibody response started after this dog was 
vaccinated witii a modif ied-li ve canine distemper /hepatitis 
vaccine at four weeks of age. The immunization of dogs 
with this vaccine has been shown to enhance antigen- 
specific IgE response to an antigen administered at the 
same time (78). It has been hypothesized that this effect 



57 

is the result of a suppression of T-suppressor cells. 
Because the mechanism that would normally suppress IgE 
synthesis is altered, IgE antibody response will develop. 
This alteration in the suppressive network and subsequent 
IgE antibody synthesis has been called the "allergic break- 
tiirough" (79). 

Surrenary 

Twenty-eight dogs imitiunized to DNP-ASC at birth and 
then three times at two week intervals produced serum anti- 
DNP antibody. The IgM response was detected one week after 
primary immunization and lasted for up to five weeks. The 
IgE and IgG antibody response in general was not present 
until week three but persisted through the immunization 
schedule. Although variation in the level and duration of 
the antibody response was detected between individual dogs, 
each dog did have a response that included all three 
i so types examined. 

Conclusions 

(1) Dogs immunized with DNP-ASC develop a high 
level, long term IgE and IgG antibody response but the IgM 
response followed a different kinetic pattern in that it 
did not persist after week five. 



58 

(2) There a difference in the responsiveness to 
this antigen seen between individual dogs for all antibody 
isotypes. This vvas inost probably a reflection of the 
genetic heterogeneity between these dogs. 



CEmPTER THREE 
MTEMPIS TO REGULATE AN ANTIBODY RESPONSE 
WITH AUTOLOGOUS ANTIBODY 

Introduction 

The rnechanisitis by v\*iich antibody responses are regu- 
lated have been stii:iied extensively. Many experiments have 
shown that antibody can be self-regulating (80,81). There 
are at least two different vays that this can occur: 1) 
If antibody is present at the time of immunization, anti- 
body can bind to and remove antigen. Therefore, the result 
would be a decrease or a failure to mount the response. 2) 
Antibody can induce an ant i- idiotypic immune response which 
vould regulate the subsequent expression of the antibody 
through id/anti-id interactions (30-82). 

If the synthesis of IgE antibody could be suppressed 
with antibody, such therapy may be very beneficial in con- 
trolling IgE nediated allergic disease. As discussed in 
Chapter one, mssively administered antibody in mice and 
rafci)its has been shown to suppress IgE antibody (25,27). 
The purpose of the experiments in this chapter is to 
determine if the administration of autologous antibody has 
any effect on the ongoing antibody response in dogs. 



59 



60 



Materials and Methods 

Affinity Chromatography 

Anti-DNP antibody vas produced by immunizing dogs to 
DNP-ASC and \-)as purified from serum by chromatography 
through a DMP-HSA affinity column as described in Chapter 
two. The bound antibody vvas eluted with 0.1 M glycine HCl, 
pH 2.5. 

RIA 

The RIA for detection of anti-DNP antibody was 
described in Chapter two. 

Aniitals and Iinmunization Schedule 

The same dogs that were described in Chapter two were 
used in these experiments. These dogs had received 100 ng 
of aluminum hydroxide precipitated DNP-ASC by the intra- 
peritoneal route on the day of birth at two week intervals 
on three further occasions. Fifteen milliliters of serum 
were obtained frcm each cfcig at the time of final antigenic 
diallenge. Anti-DNP antibody was purified from this serum 
by DNP-HSA affinity chromatography, concentrated to about 3 
rag/ml by negative pressure dialysis and rendered 
bacterially sterile by passing through a filter having 0.2 
micron sized pores. Dogs received either 10 or 100 ng of 
their own antibody anulsified in 2 ml of either complete 



61 

(CFA) or incomplete Freund's adjuvant (IFA). Dogs 
designated as controls received 2 ml of either CPA or IFA. 
In all cases, the injections were given at four sites subcu- 
taneously seven and nine weeks after the first immunization 
with DiMP/ASC. Animals were given DNP/ASC booster injec- 
tions eight and ten v«eks after the primary immunization 
(Fig.lO). 



0123456789 10 11 

A A A ^ 

o o 

= Administration of antigen 
o = Administration of adjuvant with or without 
autologous antibody 

Figure 10 
Time Schedule for Iitrounizations 



62 



Results 

DNP Affinity Column 

There was no detectable anti-DNP antibody in the 
serum of any dog after passage tlirough the DNP affinity 
column. On the other hand, the glycine HCl eluate 
contained high levels of anti-DNP IgG but no detectable 
anti-DNP IgM or IgE. Because anti-DNP IgE was not 
detected in either the effluent or the eluent from the 
affinity column but was detectable in the serum prior to 
such treatment, an aliquot of serum containing anti-DNP IgE 
was dialyzed against glycine HCl, pH 2.5 followed by 
dialysis against PBS, pH 7.2 to determine what effects 
glycine HCl had on canine IgE. There was no detectable 
anti-DNP IgE in this serum after such treatment as assayed 
by RIA. 

Iinmonization with Autologous Antibody 

As noted in Chapter two, there was considerable 
variation in the anti-DNP antibody respone between dogs. 
Tables 7 and 8 show the mean relative concentration of 
anti-DNP IgG and IgE respectively in each group of dogs 
prior to the autologous antibody administration and there- 
after. These data are presented graphically in figures 11 
and 12. The individual relative antibody concentrations 



63 



Table 7 

The iMean Relative Anti-DNP IgG Concentration 



Group 1 a) Group 2 Group 3 



N 


= 8 






N = 4 




N = 8 


Weeks 
































1 

















2 


.2 


+ 


.3 


.5+ .9 


6.5 


+ 1.7 


3 


8 .7 


+ 


7.6 


4.4 + 3.6 


13.7 


+ 7.5 


4 


12.1 


+ 


7.2 


14,3 + 6.3 


13.2 


+ 7.6 


5 


12.5 


+ 


8.4 


13.3 + 6,5 


12.2 


+ 5.6 


6 


13.7 


+ 


8.7 


12.5 + 6.0 


7.8 


+ 2.1 


7 


11.4 


+ 


7.7 


13.4 + 8.3 


10.6 


+ 5.9 


8 


12.6 


+ 


8.5 


14.2 + 8.6 


12.1 


+ 4.3 


9 


13.2 


+ 


7.9 


14.5 + 7.6 


13.8 


+ 5.9 


10 


13.2 


+ 


7.9 


14.5 + 7.6 


13.8 


+ 5.9 


11 


15.2 


+ 


9.2 


17.8 + 6.9 


15.1 


+ 7,4 



Group 4 Group 5 Control 

N = 2 N = 6 N = 8 



Weeks 
































1 







.2 


+ ,3 


.1 


+ .3 


2 


6.4 


+ 1.1 


8.1 


+ 4.4 


7.6 


+ 3.7 


3 


8.2 


+ 3.5 


12.7 


+ 3,7 


11.5 


+ 4.1 


4 


11.2 


+ 3.7 


13.7 


+ 5.2 


13.1 


+ 4.8 


5 


14,9 


+ 6.0 


15.1 


+ 7.0 


15.1 


+ 6.4 


6 


12.9 


+ 7.5 


13.4 


+ 8.9 


13.6 


+ 8.1 


7 


20.9 


+ 6.4 


16.4 


+ 7.8 


17.5 


+ 7.3 


8 


20.1 


+ 4,9 


17.3 


+ 8.5 


17.8 


+ 7.5 


9 


22.7 


+ 8.5 


18.9 


+ 7.5 


19.8 


+ 7.3 


10 


23.5 


+ 9.1 


19,4 


+ 8.0 


20.5 


+ 8.2 


11 


23.8 


+ 10.7 


17,7 


+ 6.4 


19.1 


+ 7.0 



a) All dogs received DNP/ASC Lmmunization at 0,2,4,6,8 & 10 wee]^.s, 
At 7 and 9 weeks: Group 1 received 10 [jg autologous anti-DNP 
antibody in CPA, Group 2 received 100 |ag autologous anti-DNP 
antibody in IFA, Group- 3 received 100 pg autologous anti-DNP 
antibody in CPA. Group 4 received IFA alone, Group 5 received CPA 
only. Control values were the n^an of groups 4 and 5. 



64 



Table 8 

The Mean Relative Anti-IgE Antibody Concentration 



Group 1 a) Group 2 Group 3 



Weeks 























1 


. 1 


+ .2 







.4 


+ 


.85 


2 


3.5 


+ 3.2 


2.1 


+ 3.5 


3.8 


+ 


3.4 


3 


3.7 


+ 4.3 


4.9 


+ 4.4 


10.7 


+ 


6.5 


4 


6.3 


+ 2.6 


5.2 


+ 3.6 


7.7 


+ 


1.6 


5 


9.6 


+ 4.5 


5.8 


+ 4.2 


11.4 


+ 


10.5 


6 


6.7 


+ 4.9 


3.8 


+ 3.0 


3.7 


+ 


4.5 


7 


8.0 


+ 5.3 


7.8 


+ 6.3 


10.4 


+ 


4.6 


8 


7.2 


+ 3.9 


3.7 


+ 2.7 


7.8 


+ 


2.5 


9 


8.3 


+ 3.2 


5.3 


+ 3.1 


8.8 


+ 


1.0 


10 


7.6 


+ 3.3 


4.9 


+ 2.6 


7.3 


+ 


2.4 


11 


8.3 


+ 3.8 


5.4 


+ 3.1 


7.0 


+ 


2.7 



Group 4 




Group 


5 


Control 


Weeks 




























1 















2 


2.3 


+ 1.1 


1.1 + 


.8 


1.4 + 1.0 


3 


3.6 


+ 1.3 


2.0 + 


.9 


2.4 + 1.2 


4 


3.5 


+ .7 


2.9 + 


2.3 


3.0 + 2.0 


5 


4.1 


+ 2.1 


2.6 + 


3.1 


3.0 + 2.8 


6 


2.6 


+ .4 


1.9 + 


1.5 


2.1 + 1.3 


7 


3.7 


+ 1.6 


2.4 + 


3.0 


2.7 + 2.6 


8 


6.6 


+ 2.0 


2.8 + 


1.9 


3.8 + 2.5 


9 


8.3 


+ 1.0 


3.2 + 


2.2 


4.4 + 3.0 


10 


7.1 


+ .1 


2.7 + 


2.2 


3.8 + 2.7 


11 


6.9 


+ 2.3 


3.0 + 


2.3 


4.0 + 2.7 



a) All dogs received DNP/ASC iinmuni zation at 0,2,4,6,8 and 10 
weeks. At 7 and 9 weeks: Group 1 received 10 autologous 
anti-DNP antibody in CFA, Group 2 received 100 \^q autologous 
anti-DNP antibody in JFA, Group 3 received 100 jig autologous 
anti-DiSlP antibody in CFA, Group 4 received IFA alone, Group 5 
received CFA alone. Control values were the mean of groups 4 and 
5. 



66 




uoiiDj; U90UO0 Xpoqjiuv aA!|D|9y 



0) 

■r-l 

fa 




Q 0) 



, o 0) 

ST n3 o 

O -U cu 

o 3. (0 e-' 



o 



>i (13 

>-l 4-1 

(C c 

U 8 



01 



69 



Table 9 

The Relative Anti-DNP IgE Concentration in 28 Dogs a) 



Group lb) ' (10 ng Anti-DMP Antibody in CFA) 

Animal 



Number 


1 


2 


3 


4 


Weeks 










3 


2.0+.17 


6.4+. 29 


6.1+.49 


8.1+.36 


9 


2.6+. 26 


9.3+. 84 


6.5+. 25 


9.4+. 36 


10 


3.0+.18 


8.7+. 31 


3.4+. 30 


7.1+.79 


11 


2.8+. 31 


9.3+1.26 


4 1+ 19 


8 9+ ft! 


Animal 










Number 


5 


6 


7 


8 


Weeks 










8 


5.2+. 57 


15.6+.70 


6.6+. 68 


7.2+. 30 


9 


8.9+. 63 


14.0+.69 


6.8+. 78 


9.2+1.18 


10 


9.3+. 72 


13.2+.36 


8.2+. 14 


8.2+. 07 


11 


10.2+.64 


15.1+.47 


7.4+. 66 


8.6+. 42 




Group 2 (100 uq 


Anti-DNP Antibody 


in IFA) 


Animal 








Number 


9 


10 


11 


12 


Weeks 










8 


9.8+. 47 


10.0+.22 


5.7+. 67 


5.5+. 86 


9 


9.5+1.89 


9.7+. 46 


7.5+. 83 


8.5+. 33 


10 


9.8+. 67 


8.3+. 57 


4.1+.65 


6.8+. 41 


XX 


9.4+. 59 


4.1+.12 


5.3+. 41 


9.3+. 17 


Aniiial 




Group 3 (100 ug Anti-DNP Antibodv 


in CPA) 










Number 


13 


14 


15 


16 


Weeks 










8 


2.4+. 02 


3.3+. 30 





4.2+. 21 


9 


4.1+.62 


4.0+.50 





2.9+. 14 


10 


2.0+.39 


6.3+. 35 





7.7+. 79 


11 


3.0+.48 


9.6+. 47 





2.9+. 29 


Animal 










Number 


17 


18 


19 


20 


Weeks 










8 


6.1+.22 


5.6+. 49 


.6+. 01 


7.6+. 51 


9 


9.2+. 35 


8.9+. 46 


6.0+.43 


7.0+.88 


10 


6.4+. 57 


5.2+. 34 


4.5+. 06 


6.8+. 50 


11 


7.3+. 87 


6.8+. 65 


6.4+. 06 


7.2+1.01 



70 



Table 9 Continued 



Animl 
Number 
Weeks 

8 

9 

10 
11 



Aniinal 
Number 
Weeks 



8 
9 

10 
11 



Group 4 (IFA Alone) 
21 22 



8.0+.69 
9.0+1.22 
7.1+.51 
8.5+. 34 



5.2+. 48 
7.6+. 71 
7.0+1.06 
5.3+. 44 



Group 5 Continued 
25 26 



1.0+.02 
1.9+.18 
1.1+.16 
1.4+.21 



4.3+. 34 
4.3+. 31 
3.1+.41 
2.9+. 24 



Group 5 (CPA Alone) 
23 24 



2.2+. 29 
2.2+. 20 
1.8+.36 
2.6+. 09 



27 

4.2+. 44 
4.2+. 35 
5.7+. 26 
5.2+. 38 



.3+. 02 







28 

5.0+.28 
6.3+. 06 
4.7+. 71 
5.9+. 29 



a) The level of anti-DNP IgE in these dogs from to week 7 is found 
in Table 2. 

b) Each dog received DiSlP/ASC immunization at weeks 0,2,4,6,8,10. At 
v^ks 7 and 9: Group 1, 10 \xq autologous anti-DNP antibody in CFA; 
Group 2, 100 ng autologous anti-DNP antibody in IFA; Group 3, 100 
Hg autologous anti-DNP antibody in CFA; Groip 4, IFA alone; Group 
5, CFA alone. 



71 



Table 10 

The Relative Anti-DNP IgG Concentration in 28 Dogs a) 



CFA) 
Anioal 
Number 
Weeks 

8 

9 

10 
11 



AnirtHl 
Number 
Weeks 



Group 1 b) (10 ^ig Anti-DNP Antibody in 



8 
9 

10 
11 



Aniiml 
Number 
Weeks 

8 

9 

10 
11 



Animal 
Number 
Weeks 

8 

9 

10 
11 



9.0+.11 
9.0+.66 
10.2+.93 
9.1+1.04 



5.5+. 23 
4.6+. 68 
5.8+. 89 
8.9+. 24 



24.7+. 45 
27.4+2.50 
28.5+. 76 
26.8+. 39 

Group 1 Continued 

6 

24.7+1.73 
27.0+.39 
29.1+2.13 
26.6+. 90 



17.1+1.37 
20.9+1.47 
22.9+1.10 
25.4+. 43 



6.6+. 86 
10.0+.52 
10.1+.77 

6.9+. 30 



9.4+1.37 
14.0+.38 
14.7+87 
12.3+.99 



8.3+. 75 
7.5+. 78 
6.2+1.05 
9.8+. 76 



8 

17.3+.75 
20.3+.72 
17.6+.48 
19.7+1.29 



Group 2 (100 lag Anti-DNP Antibody in IFA) 
10 11 12 



8.9+. 76 
8.8+. 99 
11.9+.26 
11.3+.69 



8.1+.62 
9.2+. 50 
8.7+. 61 
8.3+. 90 



14.2+.83 
16.4+.38 
15.4+1.12 
15.0+.32 



Group 3 (100 |jg Anti-DNP Antibody in CPA) 
13 14 15 16 



5.6+. 17 
10.3+.62 
12.7+.41 
12.2+.73 



13.9+.83 

19.0+.65 
20.4+.11 
20.3+2.24 



7.9+. 67 
11.8+1.12 
14.3+1.57 
14-7+.59 



16.7+.99 
16.0+.87 
15.0+1.01 
17.4+2.56 



72 



Table 10 Continued 



Animal 
Numter 
Weeks 

8 

9 

10 
11 



Aninnal 
Numter 
Weeks 

8 

9 

10 
11 



Animl 
Number 
Weeks 

8 

9 

10 
11 



Group 3 Continued 
17 18 



7.3+. 51 
5.5+. 60 
8.4+. 62 
9.1+.76 



25.5+3.14 
21.6+1.19 
24.3+. 29 
23.1+1.79 



Group 4 (IFA Alone) 
21 22 



15.6+.93 
16.7+.81 
16.2+1.3 
17.1+.43 



23.3+1.87 
28.7+1.61 
31.4+1.05 
30.0+2.62 



Group 5 Continued 
25 26 



9.3+. 81 
10.4+.65 
11.7+.88 
11.2+.76 



13.6+.55 
18.7+1.17 
20.5+1.48 
13.0+1.42 



19 

24.8+. 64 
29.0+2.62 
30.1+4.71 
30.8+1.06 



20 

14.6+1.73 
16.8+.96 
17.2+2.33 
22.9+1.87 



Group 5 (CFA Alone) 
23 24 



30.1+1.74 
27.7+. 50 
31.9+3.81 
26.5+2.62 



27 

20.2+.96 
22.4+. 68 
17.3+1.24 
18.7+.48 



8.1+.93 
9.5+. 62 
10.7+1.08 
9.6+. 27 



28 

22.2+1.43 
23.4+. 93 
24.3+. 74 
22.0+2.38 



b) Each dog received DNP/ASC immunization at weeks 0,2,4,6,8,10. At 
weeks 7 and 9: Group 1, 10 ng autologous anti-DMP antibody in CFA; 
Group 2, 100 jjg autologous anti-DNP antibody in IFA; Group 3, 100 
ng autologous anti-DNP anti'oody in CFA; Group 4, IFA alone; Group 
5, CFA alone. 



73 

for IgE and IgG for each dog is given in table 9 and 10. 
There was a gradual increase in tlie mean antibody concen- 
tration in general for both IgE and IgG antibody whereas 
IgM was not detected after week 5. In two dogs (15, 24), 
after the seventh and eighth weeks respectively, there was 
a cessation of the IgE response. Although the rean IgE 
antibody response increased for the groups in general, 
individual dogs varied considerably. For exairple, dogs 14 
and 18 had a peak IgE antibody response at week 7 and there- 
after the response diminished, ^^Aiereas the peak response 
for dogs 19 and 27 occurred at the end of the schedule. 

There was no marked difference in tlie antibody 
response between the different groups of dogs. To deter- 
mine if there were any patterns in the antibody response 
between these groups, an analysis of variance comparing 
time by groi:p was calculated for each antibody class with 
the assistance of the Department of Biostatistics , College 
of Medicine, University of Florida. There was no signif- 
icant difference between these groips at any given time by 
this analysis (p greater than .05). 

Because of the large variation between dogs, an 
analysis of variance vas calculated comparing dogs within a 
single litter in one group to dogs from the same litter in 
other groups as a function of time. This analysis was used 
to determine if there was variation between one treatment 
in tlie IgE or IgG antibody response as compared to a second 



74 

treatment within a single litter. In no case was a signif- 
icant difference observed. 

Discussion 

The fact that anti-DNP antibody could not be detected 
in the effluent fron tlie affinity column indicates that the 
column m.s effective in rotraving all anti-DNP antibody. 
The inability to detect IgE in the glycine eluate vvqs 
expected because canine IgE is not stable at low pH. 
Halliwell (7) has shown that at a pH of 2.5 for 30 minutes 
there is greater than a tenfold decrease in detectable IgE 
antibody. 

There are at least four possible reasons why autol- 
ogous antibody administration failed to regulate the anti- 
body response in these dogs as had been achieved in labora- 
tory animals. Firstly, in these experiments the dogs had 
an established antibody response whereas in many of the 
experimental systems \vhere passive antibody showed regu- 
latory effects on antibody production, a primry or early 
secondary response was rtHnipulated. It has been shown that 
it is much more difficult to manipulate a preexisting and 
established response than to alter a developing one. 
Secondly, there may be something unique about the regu- 
latory effects of passive antibody on an immune response in 
young animals. Antibody is transferred from nother to 



young both before and shortly after birth. If this passive 
antibody wuld result in long term suppression, then it 
might result in a negative selection process in those 
animals by rendering them immunologically non-responsive to 
pathogenic agents. Therefore, very young aniixals may be 
less susceptable to the regulatory action of passive 
antibody. In fact, in a very recent report by Jarrett and 
Hall (83), they demonstrated that maternal antibody or 
passively administered antibody given to newborn rats 
resulted in an enhanced IgG antibody response viien 
challenged with antigen at six weeks of life. However, not 
every rat so treated had this enhanced IgG response and 
scsne rats had a decrease in their IgE response. Thirdly it 
has been hypothesized that one way in which passive anti- 
body administration could regiilate antibody was through the 
generation of an anti-id response (30). It is possible 
that any anti-id produced by these cfcigs was not sufficient 
to regulate antibody. Lastly, such regulation may result 
in a clonal escape mechanism. Pawlak et al. (84) has sihown 
that the administration of anti-id to A/J mice against the 
major cross reactive idiotype produced in these mice immu- 
nized to p-azobenzenear senate would suppress this idiotype 
and other cross reactive idiotypes, but tiiere was a compens- 
atory increase in other idiotypes not ordinarily expressed 
in these mice. If the administration of autologous 
antibody had regulated a subpopulation of antibody 



76 

molecules but in response other antibody rnolecules were 
expressed, the net effect may not be observable if the 
entire class specific response were to be measured as was 
the case in these experiments. 

Stjnmary and Conclusions 

The administration of autologous anti-Dl^3P antibody in 
adjuvant to dogs which had an ongoing anti-DNP antibody 
response did not have a significant effect on this antibody 
response as compared to control dogs who received adjuvant 
without autologous antibody. This would suggest either 
that the regulation by passive antibodies, as seen in 
laboratory animals, does not cerate in this species or 
that any regulation that occurred by such treatment could 
not be detected by the methods used in this study. 



CHAPTER FOUR 
THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY 

Introduction 

Anti-idiotypic antiiDody has been produced by immu- 
nizing an animal witii isologous or autologous antibody in 
adjuvant ( 40-42 ) . The use of isologous or autologous anti- 
body rather than homologous antibody eliminates tlie poten- 
tial that allotypic determinants might be recognized rather 
than idiotypic determinants. The immunization schedule 
used in the previous experiments included the admini- 
stration of autologous antibody in adjuvant. It vss hoped 
that this treatment would regulate IgE antibody, but unfor- 
tunately, it did not. It >as not known if this failure was 
because of a lack of an anti-id response or for other 
reasons. This treatment my have induced anti-id. It is 
also possible that anti-id may have been produced during 
the immunization with antigen. The purpose of the experi- 
ments in this chapter was to determine if, at any time 
during the iimunization schedule, anti-id was detectable. 



77 



Materials and Methods 



78 



Antisera 

The antisera used in these experiments were described 
in Chapter two. Any cross reactive anti-nr>use 
immunoglobulin activity that was present in the anti-canine 
IgG, IgM or IgE antisera used in tlie anti-id RIA vvas 
reinoved by passage through an affinity column which had 
bound to it a 40 percent saturated ammonium sulfate 
precipitate of normal mouse serum. 

Animals and Immunization 

The animals and immunization schedules have been 
described in Chapter two and three except tliat in the 
experiment designed to determine if the specificity of the 
antibody was important in the induction of an anti-id 
response, a different immunization protocol was used. 
Eight rtature dogs were injected with 100 ng of aluminum 
hydroxide precipitated DNP-ASC by the intraperitoneal route 
on the day of arrival. At the sarne time, these dogs 
received a second injection of 100 ng of aluminum 
hydroxicfe precipitated ABA-KLH by the same route at a 
different site. These dogs were immunized three times at 
two week intervals. At the tin^ of the last immunization, 
30 ml of blood were obtained from each dDg. The serum fron 
tliis blood was used to purify antibody. The dogs were 



79 

arbitrarily placed into one of three groups. Group 1 (iS^3) 
received 100 jag of autologous anti-DNP antibody in CFA by 
the subcutaneous route. This antibody was purified from 
serum by adsorption to and elution from a DNP coupled 
affinity column followed by passage through an ABA cotpled 
affinity column to ensure that the piarified anti-DNP anti- 
body had no cross reactive anti-Affit antibody. Conversely, 
the autologous anti-ABA antibody in CFA for Group 2 (N=3) 
vas purified fran serum by adsorption to and elution from 
an ABA affinity column followed by passage through a DNP 
column. The control group. Group 3 (N=2) received CFA 
without autologous antibody. This later immunization was 
administered six weeks after the primary injection of 
antigen. Two weeks after tiiis last injection, serum from 
each aninial vias assayed for ant i- idiotypic antibody. 

Ant i- id RIA 

The RIA used to detect ant i- id was performed essen- 
tially as described for tlie antigen specific RIA using a 
number of mouse monoclonal anti-DNP antibodies as the 
antigen. These were 

a) Anti-DNP IgG, (a gift from Dr. A.P. Lopes, Univer- 
sity of Pennsylvania) and as a control antibody, anti-H^K 
IgG (a gift from Dr. P. Klein, University of Florida). 

b) Anti-DNP and IgE as a control antibody, anti-OVA 

IgE. 



80 

c) Anti-DNP IgM and as a control antibody, anti-SRBC 

IgM. 

The antibodies b) and c) were obtained from Sera- 
Labs, Accurate Chemical and Scientific Corp., Westbury, 
N.Y. 

d) Anti-DNP IgM (a gift from Dr. C.W. Clem, Miss- 
issippi State University. This antibody will be designated 
anti-DNP IgM-^), and as a control antibody, anti-SRBC IgM 
(a gift from Dr. W.C. Raschke, La Jolla Cancer Research 
Foundation , Ca . ) . 

The wells v^re coated with 10 tag antibody in 50 ^1 
of Tris buffer (0.1 M, pH 8 .0) . A radiolabelled 
anti-canine IgG antibody was used as the radiolabelled 
probe unless otherwise stated. 

Hapten Inhibition of Id/anti-id Interaction 

The RIA using mouse anti-DNP IgG nnnoclonal antibody 
or control IgG monoclonal antibody performed as prev- 
iously described except that after blocking any ranaining 
active sites by the addition of HSA to the plates, various 
amounts of 2 , 4-dinitrophenol glycine (Sigma Chemical Co., 
St. Louis, Mo.) ranging from 0.001 to 0.1 mg in PBS were 
incubated for three hours at 4°C. Serum samples were 
then added to the wells, incubated for three hours at 4°C 
and vsashed to remove unbound protein. A radiolabelled 
anti-canine IgG antibody was added, incubated for three 



SL 

hours at 4°C and t±ie wells were vrashed to remove unbound 
radiolabelled antibody. The anraunt of radioactivity asso- 
ciated with the well was determined in a gamma counter. 

opm sample in the presence of hapten 
% Inhibition = cpm sample in the absence of hapten X 100 

Inhibition of Antigen Antibody Interactions by 
Ant i- Idiotypic Antibodies 

The RIA using mouse anti-DNP IgG monoclonal antibody 
or a subtype and allotype imtched control crouse IgG irono- 
clonal antibody was performed except tliat after blocking 
remaining active sites by the addition of HSA to the 
plates, serum with or without anti-id was added, incubated 
for tliree hours at 4°C and vsshed three times with RAST- 
buffer. Antigen (125 I DNP-HSA, approximtely 20,000 cpm) 
was added to the wells and incubated for tliree hours at 
4 C and each well was washed five times to remove unbound 
antigen. Any ant i- id that vvas bound to the anti-DNP anti- 
body may inhibit tiiis antigen-anti-DNP interaction. The 
arraunt of antigen bound to tiie wells was determined in a 
Packard gamcta counter. The percent inhibition by anti-id 
vias calculated by 

% Inhibition = 
cpm bound to plates after serum incubation 
1- cpm bound to plates after PBS incubation X 100 



82 



Results 

Identification of Canine Anti-Iditoypic Antibody 

A screening procedure was used to assay for the 
presence of anti-id in serum obtained during the immun- 
ization schedule. Those dogs that received autologous 
antibody produced an antibody which would bind to irouse 
monoclonal anti-DNP IgG as seen in table 11. This binding 
vas not detectable prior to such treatment in any dog nor 
could it be detected in control dogs at any time. There 
vas no detectable binding to tine anti-H^K IgG nouse anti- 
body in serum from any dog. 

The experiment was repeated with another group of 12 
dogs and the anti-id activity was converted to an arbitrary 
relative antibody concentration by interpolation from a 
scale derived fran the titration of a positive high titer 
sample identified in the screening procedure (table 12). 
Ihis standard serum vas given a relative antibody concen- 
tration of 10. Some dogs produced detectable levels of 
this anti-idiotypic antibody within one week after autol- 
ogous antibody administration whereas other dogs took three 
weeks to develop such a response. There was also variation 
in the magnitude of the response observed. This anti- 
idiotypic antibody could not be detected if anti-canine IgE 
or IgM antisera vas used as the radiolabelled probe rather 



83 



Table 11 

Screening for Canine IgG Anti-Idiotypic Antibody by RIA 
Using Mouse Monoclonal Anti-DNP IgG as the Antigen 



Animal # a) 2 3 4 6 14 

Week ' ■ 






476+16 


331+19 


448+7 






1 


453+7 


590+1 


481+11 


421+10 


470+47 


2 


412+21 


347+3 


274+3 


335+6 


443+15 


3 


317+9 


277+11 


352+9 


371+3 


367+40 


4 


305+15 


358+3 


367+27 


401+16 


457+12 


5 


378+31 


376+7 


358+18 


421+11 


352+25 


6 


396+14 


417+37 


284+10 


379+23 


318+14 


7 


421+27 


335+10 


409+15 


522+27 


314+24 


8 


1357+31 


1815+26 


433+21 


2173+68 


277+43 


9 


2140+150 


2027+36 


1533+117 


2250+137 


1420+48 


10 


1862+17 


2208+55 


2092+130 


2297+100 


2 13 5+18 ( 


11 


2174+79 


3058+121 


2107+41 


3375+46 


5087+16' 


Animal # 


17 


19 


20 


23 


24 


Week 















476+1 


357+40 


307+22 


422+9 


483+19 


1 


453+17 


351+26 


406+19 


470+16 


517+42 


2 


418+5 


481+56 


464+3 


394+7 


347+5 


3 


519+43 


435+13 


253+13 


367+14 


318+23 


4 


442+16 


467+24 


373+20 


334+17 


351+1 


5 


376+18 


430+45 


481+6 


442+26 


235+7 


6 


340+28 


398+1 


295+30 


462+11 


434+16 


7 


315+7 


384+93 


563+38 


371+31 


346+34 


8 


793+13 


1937+179 


721+37 


529+21 


471+18 


9 


973+10 


2476+88 


936+68 


315+8 


512+46 


10 


1611+86 


2720+9 


2437+177 


447+60 


341+28 


11 


1735+122 


2925+66 


2026+69 


473+3 


429+27 



a) All (togs were immunized with DNP-ASC in adjuvant at weeks 
0,2,4,6,8 and 10. At weeks 7 and 9, dog 2,3,4 and 6 received 10 pig 
autologous anti-DNP antibody in CFA; dogs 14,17,19 and 20 received 
100 |ig autologous anti-DNP antibody in CFA; dogs 23 and 24 received 
CFA alone. 

b) This represents tiie mean + standard deviation of a triplicate 
sample. All sairples were' assayed at a serum dilution of 1/5 in PBS. 

The mean + standard deviation for all samples assayed using mouse 
anti-H K IgG, was 477+97. 



84 



Table 12 

The Relative Antibody Concentration of Canine IgG 
Anti-Idiotypic Antibody as i^feasured in a RIA using 
Mouse Monoclonal Anti-DNP IgG as the Antigen a) 



Aniinal 
Number 


1 


5 


7 


8 


13 


WeeJcs 






























1 

















2 

















3 

















4 

















5 

















6 

















7 

















8 


3.1+.2 


4.0+.2 











9 


5.3+. 4 


5.4+. 4 


3.1+.3 


3.1+.4 





10 


4.3+. 3 


5.1+.4 


3.8+. 2 


5.4+. 6 


3.4+. 2 


11 


5.2+. 4 


8.0+.6 


2.9+.1 


5.5+. 2 


6.6+. 4 



Animal 



Number 


15 


16 


18 


26 


27 


Weeks 






























1 

















2 

















3 





. 











4 

















5 

















6 

















7 

















8 





2.4+.1 


1.2+.1 








9 


3.7+. 2 


2.2+. 2 


2.8+.1 








10 


10.0+1.3 


4.3+. 3 


5.2+. 4 








11 


9.2+. 7 


3.3+.1 


7.1+.3 









a) The relative antibody concentration + range vas determined by 
interpolating from tlie titration of a serum sample containing high 
levels of anti-id. A value of zero indicates no detectable 
anti-idiotypic antibody. 

b) All animals received DNP-ASC adjuvant at weeks 0,2,4,6,8,10. At 
weeks 7 and 9 all animls received autologous antibody in adjuvant 
except 26 and 27 v^o received adjuvant alone. 



85 

than the anti-IgG antisera, indicating that it vas of the 
IgG class. 

When three different anti-DNP nDnoclonal antibodies 
and three control monoclonal antibodies were used, the 
binding activity \vas detected only with the original anti- 
DNP IgG (table 13), and to a lesser extent, to anti-DNP 
IgM-2 ( table 14 ) . in those animals in which anti-DNP 
IgM-2 binding activity was detected, a comparison was 
made between the serum from a point in time prior to autol- 
ogous antibody administration and serum obtained after such 
treatment. A minimum value that was two standard devi- 
ations above the irean of the control was considered indi- 
cative of anti-id activity. As was the case with the IgG 
antibody, only those animals which received autologous anti- 
body, showed binding activity and only after administration 
of autologous antibody (table 14). 

Ihe Role of Antibody in the Specificity of the 
Anti-Idiotypic Production 

To determine if immunization with an antibody Wiose 
specificity v»as other than anti-DNP would result in anti- 
DNP/anti-id, eight dogs were given both DiSfP/ASC and ABA/KLH 
three times at two week intervals. Six weeks after the 
primary injection of antigen, three dogs received 100 |jg 
of autologous anti-DNP antibody in CFA, and a different 
three dogs received 100 pg autologous anti-ABA antibody in 



86 



Table 13 

Detection of Canine IgG Anti-Idiotypic Antibody 
by RIA With Various i^use Monoclonal Antibodies 



Aninal # 
26 
27 
16 
15 
8 
7 

13 



Anti-DNP 
(IgE) 

234+7 

281+1 

171+3 

321+31 

261+13 

287+69 

389+11 

108+45 



Anti-OVA 
(IgE) 

267+25 

335+89 

286+48 

261+31 

284+14 

221+8 

205+34 

105+34 



Anti-DlsiP 
( IgM) 

264+4 

151+6 
286+3 
301+29 
322+17 
361+37 
257+28 



Animal # 
26 
27 
16 
15 
8 
7 

13 



Anti SRBC 
(IgM) 



182+36 

197+12 

176+42 

106+7 

106+7 

114+8 

189+15 



Anti-DNP 
(IgG) 



321+95 
354+29 
4777+65 
7571+246 
1273+93 
980+47 
1144+68 



Anti-ABA 
(IgG) 



209+47 

21S+38 

207+7 

176+12 

178+31 

163+14 

221+8 



SerujTi was diluted 1:2 in PBS. All samples were from week 10 of the 
LTimunization schedule. All aninals had received DNP-ASC in adjuvant 
at week 0,2,4,6,8,10 and at weeks 7 and 9 received autologous 
antibody in CFA except 26 and 27 who received CFA alone. 



87 



Table 14 

Detection of Canine IgG Ant i- Idiotypic 
Antibody by RIA with Mouse Monoclonal 
Anti-DNP IgM Antibody 

Dilution of 

Serum Anti-DNP IqM^Antibody 





26-E 


26-L 


2 7-E 


2 7-L 


1 fi-P 

J- VJ J_i 


1 f^—T 
XD i-j 


1/5 


162+11 


286+21 


135+14 


248+45 


90+7 


643+31 


1/10 


82+14 


154+15 


97+26 


152+29 


86+ 


296+19 


1/20 


71+21 


135+9 


126+36 


181+11 


103+11 


156+12 




15-E 


15-L 


8-E 


8-L 


7-E 


7-L 


1/5 


225+31 


912+25 


216+47 


570+46 


249+1 


672+6 


1/10 


184+8 


442+27 


128+6 


373+31 


125+8 


343+37 


1/20 


115+12 


247+6 


102+25 


270+12 


86+7 


236+24 




5-E 


5-L 










1/5 


123+33 


741+54 










1/10 


106+7 


358+54 










1/20 


134+8 


229+11 











a) The E indicates serum obtained prior to tlie administration of 
autologous antibody administraion (16,15,8,7) or adjuvant alone 
(26,27), at week 6. 

The L indicates serum obtained after such treatment from week 10. 
The mean + standard deviation for all samples assayed using 
anti-SRBC IgM was 186+53. 



88 

CFA. 'TwD dogs received CFA alone. Serum before and two 
weeks after this treatment vias screened for anti-id 
activity. As seen in table 15, the dogs that received 
autologous anti-DNP antibody produced an anti-id which was 
detected by the anti-DNP IgG mouse monoclonal antibody and 
failed to bind to the anti-ABA IgG. Dogs that received 
anti-ABA antibody in adjuvant developed anti-id which bound 
to anti-ABA mouse monoclonal IgG but failed to bind to the 
anti-DNP IgG mouse monoclonal antibody. The two control 
dogs produced no detectable anti-id that was reactive with 
either mouse monoclonal antibody. 

Hapten Inhibition and Elution Studies of the Id/anti-id 
Interaction 

The anti-idiotypic RIA was used to determine if 
hapten could inhibit the binding of anti-id to the mouse 
anti-DNP antibody. In two of the six cases (5,15), 10 ng 
of hapten ^^s able to inhibit id/anti-id interaction (table 
16) as shown by a slight decrease in the cpm bound to the 
wells as compared to the same sanaple incubated with PBS (22 
percent and 26 percent inhibition respectively). As the 
concentration of hapten decreased, so did the percent 
inhibition, (17 percent and 5 percent at a fivefold 
decrease in hapten concentration ) . However , it is unclear . 
how significant this inhibition was because of the 
extraiiely large amounts of hapten required to obtain these 



89 

results. Similarly, hapten was unable to elute anti-id 
from id v*en a concentration of DNP-glycine of up to 20 ng 
vas used. 

Inhibition of Antigen Binding to Antibody by 
Anti-Idiotypic Antibody 

Since hapten could not consistently interfere with 
id/anti-id, it vas reasoned that perhaps a large hapten 
coipled molecule might interfere with this interaction. 
Although the anti-id was not binding to id determinants 
within the antigen binding site, it nay have bound to deter- 
minants close enough to the hypervariable region of the 
antibody molecule to sterically hinder antigen binding to 
antibody. Serum containing anti-id was preincubated with 
anti-DNP mouse monoclonal antibody prior to tlie addition of 
a radiolabelled DNP-HSA antigen to determine if tlie 
presence of anti-id could inhibit the anti-DNP antibody 
DNP-HSA antigen interaction. As seen in table 17, anti-id 
was able to inhibit the binding of the radiolabelled 
antigen to the mouse monoclonal antibody. Animl 26 had ro 
detectable anti-id, and serum from this dog failed to inter- 
fere with antigen binding to antibody. In contrast, the 
other dogs had detectable anti-id and inhibited this inter- 
action from 21 to 50 percent of the naximum cpm bound. 



90 



Table 15 

Specificity of Cinine IgG Anti-Idiotypic Antibody 
After the Administration of 
Autologous Antigen-Specific Antibody 

Anti-ABA Anti-DMP 
Group 1 a) Group 2 b) 

Antibody in well Antibody in well 

Animal Serum Anti-DNP Anti-ABA Aniraal Anti-DNP Anti-ABA 



1 


Dil. 


I2G^ 


13^1 


# 




IgG-^ 


29 


1/5 


391+36 


1901+101 


33 


1243+37 


134+53 




1/10 


236+13 


1675+118 




996+115 


142+13 




1/20 


153+14 


1145+141 




703+10 


249+86 


30 


1/5 


293+114 


1867+157 


34 


1221+101 


113+55 




1/10 


185+26 


1383+57 




683+6 


261+69 




1/20 


85+37 


1061+46 




545+59 


143+19 


31 


1/5 


383+35 


1070+96 


35 


902+36 


317+29 




1/10 


248+89 


899+18 




531+106 


131+16 




1/20 


103+15 


494+85 




331+21 


129+21 


Control c) 












32 


1/5 


172+16 


246+42 


36 


238+33 


186+26 




1/10 


158+47 


117+28 




186+32 


133+58 




1/20 


98+8 


88+14 




121+13 


94+17 



Each dog received three L-miunizations with ABA-KLH and DNP-ASC at two 
week intervals. Six weeks after primary immunization, dogs received 
either . 

a) Group 1 received 100 ^g autologous anti-ABA antibody in CPA 

b) Group 2 received 100 pig autologous anti-DNP antibody in CFA 

c) Control received CPA alone 

d) I"5ean cpra of the sample assayed in triplicate + standard 
deviation. The serum was diluted 1/5 with PBS. 

The serum used in this assay was obtained 2 weeks after this later 
immunization. 



91 



Table 16 

The Inhibition of Canine IgG Anti-id Binding to 
iVJouse Monoclonal Anti-DiSIP IgG by Hapten 
as Measured by RIA 



2, 4 DiSlP Glycine 

|ig 
10 |ig 
2 (ug 

1 lag 
0.1 ng 



1 

3943 + 73 b) 
3802 + 100 
3983 + 280 
3807 + 214 
3804 + 12 



Animal Number a) 
5 

4522 + 138 
3533 + 69 
3733+136 
4268 + 56 
4623 + 219 



7 

1775 +34 
1704 + 91 
1675 + 46 
1614 + 101 
1734 + 73 





10 ^g 
2 ng 
1 ^g 
0.1 pg 



2281 + 162 
2203 + 200 
2012 + 115 
2148 + 83 
2213 + 129 



15 

2702 + 131 
2026 + 168 
2593 + 173 
2738 + 57 
2694 + 89 



16 

1643 + 52 
1454 + 15 
1691 + 14 
1377 + 34 
1526 + 31 



a) Each dog \^a3 immunized with autologous anti-DNP antibody in CFA 
at ^eks 7 and 9. All samples were from wedc 9 in the schedule 
except sample 15 which was from week 11. 

b) This is tlie mean cprn of triplicate samples + standard 
deviation. All samples were assayed with serum diluted one to four 
with PBS. 

A control mth mouse anti-H„K IgG, rather than anti-D^]P 

IgG was assayed for each sample. The mean + standard deviations 
for all samples 283+47. 



92 



Table 17 

Inhibition of binding of 125 I-DNP/HSA to 
Mouse .Monoclonal Anti-DNP Antibody 
by Canine Ant i- Idiotypic Antibody 



a) C.P.M. Bound + S.D. b) % Inhibition c) 

26 994 +43 

8 493 + 38 50 

18 544 + 25 45 

13 783 + 41 21 

15 611 + 13 39 

1 521 + 18 48 



a) Animal 8, 18, 13, 15 and 1 received autologous antibody in CFA 
and had detectable levels of anti-id; aninnl 26 received CFA alone 
and did not have detectable anti-id. all serum was obtained at 
week 11. 

b) This is the mean cpn + standard deviation of a sample assaved 
m triplicate ~ f j 

c) 

% Inhibition = c.p.m. control - c.p.m. sample 

c.p.m. control 

The control ^s a set of wells coated with irouse anti-DNP IgG and 
incubated with PBS rather than serum. The value for this was 
1026+19 . 



93 



Disctjssion 

Dogs immunized with autologous antibody produced an 
antibody v^ich bound to one. anti-DNP mouse monoclonal IgG. 
Hhis antibody was present only after such treatment and not 
present prior to the administration of autologous antibody. 
The specificity vas limited to id determinants present on 
some but not all anti-DNP mouse monoclonal antibodies. It 
could not be detected using mouse monoclonal antibodies 
whose specificity was other than DNP such as anti-H^K IgG 
or anti-Am IgG. This putative anti-id had no specfificity 
for mouse immunoglobulin heavy or light chain constant 
region determinants as indicated by the failure to detect 
any activity vhen an allotype and isotype n^tch 
non-anti-DNP antibody was used in the assay as antigen. 
These findings suggested that this mouse binding protein 
vas anti-idiotypic in nature. 

When the serum which contained this anti-id was 
assayed to determine if this antibody could bind to other 
mouse anti-DNP monoclonal antibodies, there was no detect- 
able binding to two of the mouse anti-DNP monoclonal anti- 
bodies and a limited binding to the monoclonal anti-DNP IgM 
antibodies. The difference in the level of anti-id 
detected when either the anti-DNP IgG or the anti-IgM anti- 
body vias used as the antigen indicates the difference in 



94 

the ability of the anti-id to bind to tliese two antigens. 
This difference could be a function of 1) different idio- 
typic determinants present on the two antibodies, or, 2) 
the difference in the accessability of tlie id to the 
anti-id or 3) a combination of both of these. 

Id/anti-id interactions can, in nany cases, be inhi- 
bited by hapten. If the interaction is hapten inhibitable, 
it suggests that anti-id binds to id determinants within 
the antigen combining portion of an antibody nolecule or to 
idiotypes intimately associated with this region. In those 
instances v^*lere hapten is unable to inhibit this inter- 
action, it can be concluded that anti-id is binding to 
those ids not within the antigen binding site. It also 
indicates that anti-id does not act as an internal image of 
antigen. High concentrations of hapten relative to the 
amount of antibody on the plate were preincubated with 
mouse anti-DNP IgG antibody. The presence of hapten did 
not consistently inhibit canine anti-id/rroiise id inter- 
action. Only two of the six samples tested showed inhi- 
bition, with 27 percent being tiie naximum inhibition. In 
other experiments, hapten could not displace the anti-id 
frcm the mouse anti-DMP antibody. These results suggest 
that the mjority of anti-id is not binding to structures 
within the antigen binding site of the mouse nonoclonal 
anti-DNP IgG. An anti-id and an internal image of antigen 
both bind to structures within the variable regions oE an 



95 

antibody nolecule. However, an internal imge of antigen 
binds to the hypervariable regions associated with the 
antigen combining site. 

If this antibody was an internal image of antigen 
then the activity should have been detected when each anti- 
DNP antibody was used as the antigen. Also, a hapten 
should inhibit the binding of an internal image of antigen 
to the respective antibody. Since anti-id bound to only 
two anti-DNP nouse monoclonal antibodies and the id-anti-id 
interactions were not consistently inhibited by hapten, it 
can be concluded that this anti-id is not an internal itiage 
of antigen. 

Although hapten could not inhibit the id/anti-id 
interactions in all cases, indicating that the recognized 
idiotopes were not within the antigen combining sites, 
these id may be very close to the antigen combining site. 
Serum v^ich contained anti-id was assayed to determine if 
the sample could interfere with the interaction between the 
mouse anti-DNP antibody and a radiolabelled dinitro- 
phenylated antigen. In all samples containing anti-id, the 
level of antigen bound to the mouse antibody was decreased, 
although complete inhibition of this binding was not 
observed. Anti-id could consistently inhibit antibody/- 
antigen interactions. However, the id/anti-id interactions 
were not hapten inhibi table. Therefore, some of tliese 
anti-ids must bind to id determinants v*iich are close to 



96 

the antigen cxxnbining site and other anti-ids bind to ids 
that are more distant. This failure to observe complete 
inhibition could have been for at least two reasons. There 
was not sufficient anti-id in the serum to block all the 
antigen binding sites. Alternatively, the anti-id bound to 
id determinants located on the molecule in such a way that 
complete inhibition of the antigen binding site was not 
possible. 

The results in this chapter show that those aniimls 
given autologous antibody in adjuvant produced an ant i- id. 
This anti-id response is not a function of the adminis- 
tration of adjuvant because control dogs given adjuvant 
without antibody failed to produce detectable levels of 
anti-id. In those dogs that produced anti-id the autol- 
ogous antibody that was used for immunization had been 
subjected to harsh treatment (e.g. glycine HCl elution from 
an affinity column) during the purification process. This 
treatment could possibly alter tlie ids present in the anti- 
body. Therefore it might be possible that the antigen 
specificity of the antibody has little or nothing to do 
with the anti-id tliat is produced. 

To address this question, dogs were immunized to two 
different antigens, DNP-ASC and ABh-YUi. They were tlien 
given either autologous anti-DMP antibody or autologous 
anti-ABA antibody anulsified in adjuvant and subsequently 
produced an anti-id which bound to mouse monoclonal 



97 

antibody of the same specificity as the immunizing 
antibody. Prior to such treatment tiiis autologous antibody 
is present in the dog but does not induce detectable levels 
of anti-id. However, after the administration of this same 
antibody, anti-id is detected. Therefore, these results 
indicate that during the purification procedure and/or tlie 
immunization procedure, anti-id determinants present on the 
autologous antibody are immunogenically enhanced. However, 
any changes that occur in the protein molecule must be 
subtle because the specificity of the anti-id response vas 
determined by the specificity of the imiiiunizing antibody. 
If there vas narked change in the id determinants, then the 
anti-id might not be expected to maintain its specificity 
for the immunizing antibody. However, it is not possible 
from these results to determine if the immunogenic 
enhancement of the id was a result of the purification 
process, the route of immunization or a combination of both 
these things. 

It was very fortunate that the mouse antibody used as 
the antigen in these assays had id determinants which could 
bind to the canine anti-id, although shared idiotypy 
between animals has been reported ( 53 , 57 , 85-87 ) . 



Summary 



98 



The results in this chapter suggest the following: 
Anti-id can be induced by the administration of autologous 
antibody. The majority of this anti-id is not hapten 
inhibitable and is detectable with only a few tionoclonal 
antibodies of the same specificity which presumably bear 
tiie same or a similar set of cross reactive idiotypes. 
Furthermore, this anti-id is not an internal image of 
antigen . 

Conclusions 

1) The administration of autologous antibody in 
adjuvant induces a reciprocal anti-idiotypic antibody 
response. 2) The identification of these anti-id anti- 
bodies was achieved by the use of monoclonal anti-DNP 
antibody from another species as a soxarce of idiotype. 



CHAPTER FIVE 
DETSCnON OF AiNfTI-IDIOTYPIC ANTIBODY 
USILNJG AUTOLOGOUS ANTI-DNP F(ab)'2 FRAGMEbfTS 
AS THE IDIOTYPIC ANTIGEN 

Introduction 

Anti-idiotypic antibody, as noted in Chapter one, can 
have a regulatory function during an iinmune response. It 
can either enhance (62) or suppress (34,88) the level of 
the corresponding idiotype. Even if anti-id stimulates 
only a limited number of ids, the overall result is an 
enhancement in the total antigen-specific antibody response 
(63). Anti-id that acts as an internal image of antigen 
can stimulate or enhance an immune response in a way anala- 
gous to antigen (89). On the other hand, anti-id has been 
shown to suppress an entire antigen-specific isotype 
(60-61). If anti-id is important as a natural means to 
regulate antigen-specific antibody, it would sean logical 
that anti-id vrould be detectable during a normal immune 
response. The results of Chapter four indicated that after 



99 



100 

immunizing dogs with autologous anti-DNP antibody, anti-id 
was detected. However, this anti-id was not detected 
during the response to DNP-ASC. 

The failure to detect ant i- id prior to the immuni- 
zation with antibody in adjuvant could be because; 1) 
anti-id vas not present or, 2) the method used to detect it 
v^s wrong. The purpose of the experiments in tliis chapter 
were to determine if anti-id could be detected at any time 
during a DNP specific antibody response using autologous 
anti-DNP F(ab)'2 antibody fragments as the source of id. 

Materials and Methods 

Preparation and Iimiobilization of Anti-DNP F(ab)' ^ 
Fragments to a solid Matrix - 

Anti-DNP F(ab)'2 fragments were prepared from anti- 
body purified fron a single serum sample by affinity chroma- 
tography. The protein was digested with pepsin and the 
F(ab)'2 was separated from intact antibody and Fc frag- 
ments as previously described in Chapter two. The 
F(ab)'2 fron each sanple was handled separately and 
F(ab)"2 fron a single sample will be referred to as a 
set. Each set of antibody fragments was bound to a solid 
support matrix ( Immunobead^ . BioRad Laboratories, 
Richmond, CA) as described by the manufacturer. Briefly, a 
given quantity of anti-DNP F(ab) ' in 0.003 M 



101 

^2 ^^4^ buffer, pH 6.3 and a proportionate amount of 
beads v^re incubated together for one hour at 4°C 
followed by the addition of l-ethyl-3 (3-dimethylaniino- 
propyl) carbodiimde HCl (EDAC) with an additional incu- 
bation at 4°C for one hour. Any remaining active sites 
were blocked with 1 percent HSA in 0.005 phosdiate buffer, 
pi 7.2 by incubating tliis with the beads for one hour at 
roan temperature. The beads were pelleted by centri- 
fugation at 1,000 x g for 10 minutes at 4°C and alter- 
nately washed with PBS, pH 7.2 followed by 1.4 M NaCl-PBS, 
pi 7.2 three times to remove unbound protein. After the 
final wash the beads were suspended in RAST+ buffer. The 
beads used in a single experiment were standardized for 
both DNP-HSA binding and total F(ab)'2 content by incu- 
bating an aliquot from each bead set with various dilutions 
of -'I DNP-HSA or I anti-canine light chain anti- 
body. For exanple, 50 ng of one bead set bound 5,656 + 
61 cpm radiolabelled anti-canine light chain specific 
antibody (this namber of cpm is approximately 140 ng of 
anti-canine light chain antibody) and 1,040 + 53 cpm 
radiolabelled DNP-HSA. A second set bound 4,690 + 79 cpm 
anti-light chain antibody (this is approxinately 110 ng of 
anti-canine light chain antibody) . and 747 + 21 cpn 
antigen. The second set of beads had approximately 75 
percent of the binding capacity of the first set. 
Therefore 63 nl of beads fron the second set were used in 



10 2 

the assay and 50 \a1 of the first. No immuno-reactive Fc 
iraterial ^ms detectable on any bead set when "'"^^I 
anti-canine heavy chain specific IgG was incubated with an 
aliquot of each- bead set. 

Detection of Natural Occurinq Anti-id 

Anti-DNP F(ab)'2 fragments frcm a single serum 
sample immobilized as described above vas used as an 
antigen to detect anti-id. Various serum samples frcm the 
same dog were assayed for anti-id after being ciironato- 
graphed through a DNP-affinity column to remove anti-DNP 
antibody ^ich theoretically could compete by binding 
anti-id. Each sample was concentrated by negative pressure 
dialysis to approximately the starting volume of serum. 
The samples were assayed by incubating an undilute, a 1/2 
and a 1/4 dilution in PBS of each sample with a 
standardized amount of autologous anti-DNP F(ab)'2 bound 
beads for three hours at roan tanperature. The beads were 
then centrifuged and the supernatant rennoved and vjashed 
with RAST+ three tin^ to ranove unbound antibody. Radio- 
labelled heavy ciiain specific anti-canine IgG ;^as added to 
each set of beads (approximately 30,000 cpcr/ sample ) , 
incubated for three hours at roan temperature and washed to 
remove unbound radiolabelled antibody. The radioactivity 
of each sample vas determined in a Packard gamm counter. 
Included in ^ch assay at all sample dilutions were beads 



103 

bound with nornial canine IgG FCab)'^ (with no detectable 
anti-DNP activity) and with HSA boiand beads. Specific 
binding was calculated by using the following formula: 
specific binding = cpm bound to autologous anti-DNP 
F(ab)'2 beads of a sample at a given dilution - cpm bound 
to NCS IgG F(ab)'2 beads of the sample at the san^ 
dilution. Since beads were standardized for an anraunt of 
antibody in each experiment, the volume of beads used 
ranged from 50 lal to 78 \xl per sample. V\fhen 50 or 100 
lal of non-specific F(ab)'2 was incubated with the 
sample, there was less than a 15 percent difference in the 
cpm indicating that the increase in bead volume had little 
influence on the background activity. 

Results 

Identification of Anti-Idiotypic Antibody Using 
Autologous Idiotypes 

The purpose of these experiments was to determine if 
autologous id could be used to detect anti-id. Anti-DNP 
F(ab)'2 fragments from various time points in the irnnuni- 
zation schedule were used as antigens to detect anti-id in 
autologous serum. The autologous serum used in these 
experiments were first chronatographed through a DNP 
affinity column to ranove anti-DNP antibody. This was done 
to eliminate any possible interference the presence of this 



IW 

antibody might have. Ant i- id was detected during tl^ie 
DNP-ASC immunization schedule in three of the five dogs 
tested (tables 18-20) but no anti-id was evident in the 
other two dogs. The kinetics and the amount of anti-id 
varied depending upon v^at set of ids were used an antigens, 
and v^ich serum sample was tested. Three different 
patterns in the appearance of anti-id are seen: Pattern 
one: Anti-id could not be detected before or coincident 
with the id but could be detected later, as was seen with 
two samples in two dogs (figure 13,14). In dog 14, the ids 
used to detect anti-id were fron week two, anti-id was not 
detected until week seven (figure 13). Similarly, in dog 
1, v^en the ids from week four were used as antigens, 
anti-id vias not detected until week six ( figure 14 ) . 
Pattern two was seen in three dogs using six serum samples. 
In a single saiiple, id and anti-id were both present 
(figure 15-20). For example, v^en the antibody obtained at 
week six was used as an id antigen, anti-id was detected at 
week six but the mximum level of anti-id was later than 
week six (figure 15). In two samples assayed, the highest 
level of anti-id was detected fron tlie same samples that 
were used to obtain the antibody v^ich was used as the id 
antigen (figures 19 and 20). Pattern three: In dog 14, 
v*ien the ids which were used as antigens to detect ant i- id 
were fron blood obtained at week 11, ant i- id was detected 
with each sample tested (figure 21). Similarly, by using 



105 



Table 18 

The Detection of Canine Anti-Idiotypic Antibody 
by RIA Using Autologous Anti-DNP F(ab)'2 as the Id 

Dog Number 1 

Source of Ant i- id (Week) a) 
2 4 6 8 11 



Source of 
id 

(Week) b) 

Effluent 
Dilution 






82+64 


12+12 


1332+112 


2035+279 


150+48 


1/2 


41+70 


51+38 


939+99 


1519+199 


286+45 


1/4 


63+49 





171+34 


1069+170 


14+18 





37+31 


193+39 


611+50 


1241+179 


1586+42 


1/2 


14+13 


86+55 


534+37 


370+27 


1333+26 


1/4 


10+21 


3+5 


58+35 


12+25 


413+53 





3+5 


253+26 


455+37 


1496+14 


1300+17 


1/2 


43+12 


279+121 


397+61 


534+29 


714+51 


1/4 


38+27 


21+28 


179+41 


30+21 


346+15 





51+25 


426+41 


556+47 


595+257 


459+101 


1/2 


14+23 


349+153 


438+32 


437+89 


257+57 


1/4 


150+63 


79+71 


214+83 


139+81 


179+68 



Control c ) 
186+9 
1/2 149+31 
1/4 138+46 

101+7 

a) Serum from different times during the immunization schedule. 

b) The id was autologous anti-DNP F(ab)' iinmobilized to a solid 
matrix. The dog vas immunized with DNP-ASC in adjuvant at week 
0,2,4,6,8 and 10 and received 10 (jg autologous anti-DNP antibody in 
CFA at weeks 7 and 9. 

c) Control id was normal canine IgG F(ab)' iirmobilized to a solid 
matrix . 



106 



Table 19 

Detection of Canine Anti-Iditoypic Antibody by RIA 
Using Autologous Anti-DNP F(ab)'2 as the Id 

.Dog Number 14 

Source of Anti-Id (Week) a) 
2 4 7 11 

Source of 
id 

(Week) b) 

Effluent 
Dilution 



2 





137+31 


65+21 


644+68 


705+18 




1/2 


36+18 


7+7 


152+43 


351+47 




1/4 





5+7 


12+11 


90+15 


5 





216+39 


318+49 


381+68 


848+9 




1/2 


135+23 


37+31 


150+62 


677+34 




1/4 


77+64 





47+15 


300+23 


11 





624+5 


778+18 


838+52 


405+77 




1/2 


309+7 


653+28 


786+38 


233+16 




1/4 


99+29 


386+16 


493+14 


83+11 






Coptrol c ) 













321+37 










1/2 


186+7 










1/4 


128+29 









a) Serum fran different times during the immunization schedule. 

b) The id was autologous anti-DNP F(ab)' immobilized to a solid 
matrix. The dog was immunized with DNP-ASC in adjuvant at weeks 
0,2,4,6,8, and 10 and received 100 |ig autologous anti-DNP antibody 
in CFA at weeks 7 and 9. 

c) Control id was rormal canine IgG F(ab)' immobilized to a solid 
matrix. ^ 



107 



Table 20 

The Detection of Canine Anti-Iditoypic Antibody by RIA 
Using Autologous Anti-DNP F(ab'2 as the Id 

Dog Number 21 

Source of Ant i- Id (Week) a) 
1 3 5 7 10 

Source of 
id 

(Week) b) 





Effluent 












Dilution 










1 





358+94 


1000+49 


307+69 


435+146 


684+33 




1/2 


247+37 


539+97 


204+39 


197+55 


377+36 




1/4 


88+23 


102+61 


35+7 


27+26 


246+29 


3 





253+63 


1532+88 


1031+63 


299+27 


31+28 




1/2 


77+31 


865+109 


940+39 


5+3 


150+123 




1/4 


83+71 


276+123 


220+47 


2+4 


121+38 


7 





104+60 








1466+229 


1077+10 




1/2 











364+97 


591+101 




1/4 


46+29 





75+50 


169+11 


359+66 


11 





49+41 





89+31 










1/2 





65+58 


28+49 


69+63 


64+57 




1/4 


108+39 


17+11 


21+43 


101+34 


106+29 



Control c ) 
238+83 
1/2 211+61 
1/4 103+48 



a) Serum from different times during the iiranunization schedule. 

b) The id was autologous anti-DNP B"'(ab)' immobilized to a solid 
matrix. The dog was immunized to DNP-ASC in adjuvant at weeks 
0,2,4,6,8, and 10 and received CPA alone at weeks 7 and 9. 

c) Control id was normal canine IgG F(ab)' imoobilized to a .solid 
matrix. ^ 



CO 



00 
U3 



109 




1 o 

O 00 CD 'sT CM 

001 ^ ^do 



Q >i 



0) 
i-> 



0) c 

U (0 0) 

>-J CO >i 

Cn O O ^ 



C Eh 



Ill 




1 o 

CVJOOOCD^OOOOOCD^OO 
C\J CVJ 

001 X ^do 



s 



U3 U 



Em •'-I 



in 



t3 



0) T3 -H 



00 



CN 



ITS 

0) 



113 




00CD'sj-C0O00(X)'sJ-C0 



001 ^ 




■H 
I 

■H 

(13 



C 
O 

■H 
-P 

s 

-H 

U-l 
■H 
-t) 
C 



-M T3 - 

<D 3 0) — 



c 

« o 



T) m 0) t3 

55 s: 

^ 00 Eh 0) 



El, 0) 
Q 



C 
4-) 

o 



J2 CTi £ , 
S ^ O (D C 
CN TJ -H 4-) 03 



115 




iiii 
|f 'i 

I' 



Q) 

•r-l 



Cb to 



"nT J_1 -r-l 

o c J= 



(6 



V-l CD 



fa 



J 



s c 

Q (0 



T3 



•H QJ -D 



+J CO 

C CM •» 

0) >x) 

jC C ^' 

e-< o - 



119 




Figiare 19. 



The identification of anti-id in various serum 
sanples over time, in dog 24, as measured by RIA. 
The dogs received DNP-ASC in adjuvant at weeks 
0,2,4,6,8,10, and CFA at weeJcs 7 and 9. The arrow, 
narked idiotype probe, indicates the time from v^ich 
the anti-DNP F(ab)'2 fragments came. These were used 
as antigens to detect anti-id. The bars represent 
the standard deviation of the mean. 



121 




Figure 20. 



The identification of anti-id in various serum 
samples, in dog 24, as measured by RIA. The dog 
received DNP-ASC in adjuvant at weeks 0,2,4,6,8^10, 
and CFA at weeks 7 and 9. The arrow, marked idiotype 
probe, indicates the time from which the anti-DNp 
F(ab)'2 fragments come. These were used as antigens 
to detect anti-id. The bars represent the standard 
deviation of the rrean. 



123 




2 4 6 8 10 

Weeks 



0) 

a 

■H 



a. 



T3 TJ 



00 



Eh nj o cr\ 



125 




S2 2 C 
Q Q -H 

•H -H ■«. 



CP CO 



•fa 



00 



127 




001 ^ Ludo 

I 
1 



I 



0^ 



m 

CM 

(U 

3 

Cn 



129 




00 CD ^ CV) 

001 ^ ^do 



130 

ids obtained from dog 1 at week 10, anti-id was detected in 
all but the first sample tested (figure 22). Both of these 
dogs had received autologous antibody in adjuvant. Both 
these ids \vere obtained from the dogs after this treatment. 
In contrast to tlie anti-id detected in these two dogs, id 
vas obtained from dog 24 at a similar time in the immuni- 
zation schedule (week 11), but revealed no anti-id using 
these id antigens. This dog received adjuvant without 
autologous antibody. In two dogs there was no detectable 
anti-id at any point in the immunization schedule using 
autologous id as the probe. This is in spite of the fact 
tliat one of the dogs was immunized with autologous antibody 
and did have detectable levels of anti-id using the mouse 
monoclonal antibody as tlie id probe. 

Discussion 

In the experiments in Chapter four, anti-id was only 
detected in serum after autologous antibody administration. 
If anti-id does serve in a regulatory fashion, it is not 
clear why its presence is not detected during tine course of 
the anti-DNP antibody response. This inability to detect 
anti-id throughout tlie immunization schedule may, be a 
function of the xenogeneic probe used to detect anti-id or 
it may be because anti-id is present only after artificial 
manipulation. Furthermore, if anti-id does regulate the 



131 

reciprocal id, then it vvould be expected that id would be 
present in the serum tefore the appearance of anti-id, but 
after the appearance of ant i- id the reciprocal id vvould 
di;3appear. Since a number of different ids were used as 
the probe, the corresponding ant i- id iray be detected either 
slightly before the point in time the id came from, coin- 
cident with the anti-id, or considerably later in tinne than 
the id. Alternatively, if id/anti-id v^re complexed then 
the disruption of these complexes might allow anti-id to be 
detected. 

When autologous anti-DNP Ftab)'^ was used to assay 
for anti-id, three different patterns in the detection of 
anti-id were evident. In the first pattern, anti-id was 
detected after tiie appearance of id, but not coincident 
with nor before its appearance. These results suggest tliat 
there vas a lag fhase between the appearance of id and the 
corresponding anti-id in tlie serum. In the second pattern, 
id and anti-id are present within a single serum sample. 
The maximum level of anti-id was detected in serum at the 
same time that tiie id appeared in two cases. Furthernrare , 
anti-id were present very early in the response. Because 
the dogs in these experiments would still be expected to 
have colostrum-derived antibody, tiiese anti-id may 
represent maternal immunoglobulin. Unfortunately, it was 
not possible to obtain serum fron these bitches to 
determine if they had ant i- id present. 



132 

Both the first and second pattern of ant i- id are 
consistent with the hypothesis that id is acting as an 
antigen to induce anti-id. These data are raniniscent of 
the type of curves seen Wien one plots the disappearance of 
antigen as a function of time and superimposes on that 
curve the appearance of antibody that is specific for the 
antigen (90). When antigen is first introduced into an 
aninal there is initially a very slow loss of this antigen 
fran the circulation. After a few days, however, there is 
a precipitous drop in the level of antigens v*iich is the 
result of antibody production and is called iinmune elimi- 
nation. Antibody ^en first produced is not detected 
because it is cornplexed with antigen and removed fran the 
circulation. At a certain point, however, both antibody 
and antigen will become apparent in a cornplexed form. The 
variables that determine this point include the valence of 
the antigen and its size, the isotype of the antibody, the 
affinities between the antibody and the antigen, and the 
efficiency of the reticuloendothelial system in removing 
these complexes (90,91). The similarity between immune 
elimiration of antigen and the experimental results 
obtained suggest that id is ranoved in a fashion analagous 
with antigen ranoval. The appearance of anti-id in the 
tliird pattern is difficult to explain. In this case, 
anti-id was present at a time considerably before the 
appearance of id. That is, in dog 14, anti-id was 



133 

detectable in every serum sample checked (figure 21), and 
in dog 1, anti-id was detectable in all but the very 
earliest serum sample (figure 21). Both of these dogs had 
received autologous antibody in adjuvant and the idiotype 
used as the antigen was obtained from serum after such 
treatment. In contrast, when idiotypes from a comparable 
time in the Lmmunization schedule were used as antigen 
from, dog 24, which received adjuvant without autologous 
antibody, the unusual appearance of ant i- id was rat 
observed. A possible explanation would be that the 
administration of autologous anti'oody in adjuvant 
stimulated the production of antibody having similar 
idiotypes. That is, anti-DNP antibodies used for 
immunization were from week; six. The ids on this 
immunizing antibody may have stimulated additional antibody 
with the same id. Any anti-id v^iich would be produced 
because of the id of week six might also bind to ids 
produced from the immunization of the antibody from week 
six. Therefore, if this were the case, anti-id could be 
present in serum prior to tlie sample fran which the id was 
derived. There is experimental precedence for this 
suggestion (63,64). Forney et al. (63) have shown that 
mice given hybridDraa-derived anti-sheep red blood cell anti- 
body without stimulation with antigen will subsequently 
produce anti-SRBC antibody of a similar idiotypic speci- 
ficity to the Immunizing antibody. Their interpretation 



134 

vas that the antibody stimulated the subsequent production 
of identical or very similar antibody through an id/anti-id 
interaction . Therefore, this unusi:ial pattern in the 
appearance of ant i- id could be the result of autologous 
antibody administration. However, the presence of this 
anti-id may be the result of factors governing the 
production and detection of anti-id which are unforseen at 
this time. 

In two of the five animals there is complete failure 
to detect anti-id using autologous antibody in any serum 
obtained throughout the immunization schedule. There are a 
number of different possible explanations for this result. 
Firstly, only anti-id of the IgG class was iteasured and it 
is possible that other anti-id isotypes were produced in 
these two animals. In fact, in a recent study it was 
observed that in rmn there was an isotypic shift over time 
of the anti-id specific for a given set of auto-antibodies 
(53). Alternately, there may be certain ids v^ich favor 
the production of the reciprocal anti-ids. In an experi- 
irent in outbred ral±)its, anti-id production seemed to be 
associated with the presence of a few ids that favor 
anti-id production. Those rabbits not expressing such ids 
failed to produce detectable anti-id antibodies (57). 
.Based on this, it is possible that, in dogs, certain id are 
especially important for anti-id production and in those 
dogs not expressing such ids there is a failure to produce 



135 

reciprocal anti-id. This lack of anti-id woiild not 
necessarily result in abnormal antibody regulation if these 
animals had an alternate means to accanplish this, such as 
a T-suppressor cell pathway. 

One of these animals had no recognizable ant i- id 
using autologous antibody as the id probe, but did have 
recognizable anti-id when mouse monoclonal antibody was 
used as the id probe. There are several possible reasons 
for this. 1) There are only a few idiotopes present on 
monoclonal antibodies while in a heterogeneous population 
of molecules there would be expected to be many idiotypes 
and therefore even more idiotopes. Therefore, this 
negative result may be a function of the concentration of 
id present in the assay system. 2) The purification 
process may have altered the id on the antibody molecules 
just enough so that v^en this antibody was used to immunize 
a dog, these altered id were able to induce an anti-id 
response specific for the mouse id. Or 3) the ids 
detected with the mouse antibody were different than the 
ids on the canine anti-DNP F(ab)'2. 

Surtmary and Conclusions 

In three of five dogs immunized with DNP-ASC, anti-id 
could be detected using autologous anti-DNP F(ab)' 



136 

fraginents as the id. The kinetics in the appearance of 
this ant i- id in relationship to the id suggest that id is 
acting as an antigen to stimulate a corresponding anti-id 
response . 



CHAPrER SIX 
CDISJCLUSION 

Dogs were chosen to study IgE antibody synthesis and 
regulation because they develop an IgE mediated disease 
that is very similar to atopic disease of mn. 

When dogs were immunized with 100 fjg of aluminum 
hydroxide precipitated DNP-ASC by the intraperitoneal 
route, each dog synthesized anti-DNP IgG, IgE and IgM anti- 
body. There vb.s a difference in the responsiveness to this 
antigen between individual dogs vhich most probably 
reflected the genetic heterogeneity between them. In an 
attanpt to regulate anti-DNP antibody, autologous anti-DNP 
antibody in adjuvant was administered to these dogs during 
the ongoing response. Although dogs so treated did not 
have a difference in the mgnitude of the anti-DNP IgE or 
IgG response, as compared to control dogs \iiio received 
adjuvant without antibody, these dogs did produce an 
anti-id which was detected using mouse monoclonal anti-DNP 
IgG and IgM. 

This anti-id vras not an internal image of antigen as 
shown by 1) a failure of these antibodies to bind to each 



137 



138 



mouse monoclonal anti-DNP antibody tested and 2) a failure 
of hapten to inhibit the id/anti-id interaction. An 
internal inage of antigen should bind to corresponding anti- 
body to an extent similar to antigen and should also be 
hapten inhibi table (91). 

Anti-idiotypic antibody could be detected during the 
DNP-ASC immune response when the ids used to assay for anti- 
body were autologous anti-DNP F(ab)'2 fragments. These 
anti-ids can be consictered natural in that their appearance 
was associated with anti-DNP antibody produced during an 
immune response. This is in contrast to the anti-ids 
detected with the mouse antibodies. These latter ant i- ids 
were detected subsequent to the immunization with anti-DNP 
antibody in complete Freund's adjuvant. Three of the five 
dogs assayed for anti-id using autologous ids had detect- 
able amounts of anti-id whereas two dogs had no detectable 
levels. It ■m.s conclijded that anti-id can be detected 
during a DNP-ASC response in dogs. 

This method of detecting anti-id would allow for the 
ictentif ication of a different anti-id response during other 
immune responses. For example, the aim of hyposensi- 
tization for allergic disease is felt to be the production 
of a blocking antibody. However, this treatment is not 
always effective, even if blocking antibody is produced. 
An additional possible reason why hyposensitization works 



139 

is because IgE antibody is regulated by id/anti-id inter- 
action. If this is the case, the identification of ant i- id 
during hyposensitization my help in establishing more 
effective immunotherapy for allergies. 



REFERErCES 



1. Patterson, R. (1969) Laboratory Models of Reaginic 
Allergy. Prog. Allergy. 13:332-407. 

2. Patterson, R. , and Kelly, J.F. (1974) Animal I^iodels of 
the Asthmatic State. Ann. Rev. Med. 25:53-58. 

3. Pepys, J. (1975) Atopy in Clinical Aspects of 
Immunology , third ed. Edited by Gell P.G.H., Coombs, 
R.R.A., Lachmann, P.J. Blackwell Scientific Publications, 
Oxford, England. 

4. Sdiwartzman, R.M. and Rockey, J.H. (1972) Atooy in the 
Dog. Arch. Derm. 96:413-422 

5. Halliwell, R.E.W. , Sdiwartzman , R.M., and Rockey, J.H. 
(1972) Antigenic Relationship between Hunan IgE and Canine 
IgE. Clin. Exp. Irmiunol. 10:399-407. 

6. Halliwell, R.E.W. , Schwartzman , R.M. , Montgomery, P.C., 
and Rockey, J.H. (1975) Physicochanical Properties of 
Canine IgE. Transplant Proc. 8:537-543. 

7. Halliwell, R.E.W. (1972) Studies on Canine IgE . 
Ph.D dissertation. Cambridge University, Cambridge, 
England. 

8. Johansson, S.G.O. and Mellbin, T., Vahlquist, B. (1968) 
LTTOunoglobulin Levels in Ethiopian Preschool Children with 
Special Reference to High Concentrations of Immunoglobulin 
E. (IgND). Lancet 1:1118-1121. 

9. Rosenberg, E.B., Whalen, G.E., Bennich, H. and 
Johansson, S.G.O. (1971) Increased Circulating IgE in a 
New Parasitic Disease — Humn Intestinal Capillariasis . N. 
Eng. J. Med. 283:1148-1149. 

10. Tada, T., Ishizaka, K. (1970) Distribution of Gamna E 
Forming Cells in Lyrtphoid Tissues of the Human and Monkey. 
J. Immunol. 104:377-387. 

11. Jarrett, E.E.E., and Miller, H.R.P. (1982) Production 
and Activities of IgE in Helminth Infection. Prog Allergy 
31:178-233. 



14 



141 



12. Katz, p.H. (1980) Recent Studies on the Regulation of 
IgE Antibody Syntliesis in Experimental Animals and xMan. 
Inmunol . 41 : 1-24 . 

13. Ishizaka, K. (1976) Cellular Events in the IgE 
Antibody Response. Adv. Iirmunol. 23:1-75. 

14. Shinohara, N. and Tada, T. (1974) Hapten-Specif ic IgM 
and IgG Antibody Responses in Mice against a 

Thymus- Independent Antigen (DNP-Salcnonella) . Int. Arch 
Allergy Appl. Immunol. 47:762-776. 

15. Suenura, M. , Yodai, J., Hirashiiia, M. and Ishizaka, K. 
(1980) Regulatory Role of IgE binding Factors from Rat 

T- lymphocytes . 1. Mechanism of Enhancsnent of IgE Response 
by IgE-Potentiating Factor. J. Iittnunol. 125:148-154. 

16. Saryan, J.A., Leung, D.Y.M., and Geha, R.S. (1983) 
Induction of Human IgE Synthesis by a Factor Derived from 
T-cells of Patients with Hyper- IgE States. J. Immunol. 
130:242-247. 

17. Katz, D.H., Davie, J.M., Paul, W.E., and Benacerraf, B. 
(1971) Carrier Function in Anti-Hapten Antibody Responses. 
IV Experimental Conditions for the Induction of 
Hapten-specif ic Tolerance or for the Stimulations of 
Zknti-Hapten Anmnestic Responses by 'non- immunogenic' 
Hapten-Polypeptide Conjugates. J. Exp. Med. 134:201-224. 

18. Sehon, A.H. and, Lee, W.Y. (1981) Specific 
Downregulation of IgE Antibodies by Tolerogenic Conjugates 
of Haptens and Allergens with Synthetic Polymers. Int. Arch 
Allergy Appl . Itimunol . 66 ( suppl . 1 ) : 39-42 . 

19. Liu, F.T., Bogowitz, C.A., Bargatze, R.F., Zinnecher, 
M. , Katz, L.R., and Katz, D.H. (1979) Imnunologic 
Tolerance to Allergenic Protein Determinants: Properties of 
Tolerance Induced in Mice Treated with Conjugates of Protein 
and a Synthetic Copolymer of D-Glutamic Acid and D-Lysine. 
J. Iimiunol. 123:2456-2465. 

20. Zeiss, C.R. (1979) Inrounotherapy in Allergic Disease. 
In Allergy and Clinical Immunology , Medical Examination 
Publ. Co., Inc., Garden City, N.Y. 

21. Halliwell, R.E.W. (1977) Hyposensitization in the 
Treatnnent of Atopic Disease. In Current Veterinary Therapy 
VI. W.B. Saunders Co., Phila, Pa. 



14 2 

22. Faith, R.E., Hessler, J.R., Small, P.A. (1977) 
Respiratory Allergy in the Dog: Iryiuction by the 
Respiratory Route and the Effect of Passive Antibody. Int. 
Arch. Allergy Appl. Iimiunol. 53:530-544. 

23. Smith, T. (1909) Active Iitmunity Produced by 
So-Called Balanced or Neutral Mixtures of Diphtheria Toxin 
and Antitoxin. J. Exp. iVted. 11:241-256. 

24. Uhr, J.W. , and itoller, G. (1968) Regulatory Effect of 
Antibody on the Iirmune Response. Adv. Inmunol. 8:81-127. 

25. Oian, P.L., and Sinclair N.R. St. C. (1971) 
Regulation of the Irmiune Response. V. An Analysis of the 
Function of the Fc Portion of Antibody in Suppression of an 
Iimtune Response with Respect to Interaction with Conponents 
of the Lynphoid System. Immunol. 21:967-981. 

26. Strannegard, 0., and Belin, L. (1970) Si5>pression of 
Reagin Syntliesis in Rafctoits by Passively Administered 
Antibody. Immunol. 18:775-785. 

27. Tada, T., and Okumura, K. (1971) Regulation of 
Honocytotropic Antibody Formation in the Rat. 1. Feed-Back 
Regulation by Passively Administered Antibody. J. Iimiunol. 
106:1002-1011. 

28. Ishizaka, K., and Okudaira, H. (1972) Reaginic 
Antibody Formation in the iyiouse. 1. Antibody-Mediated 
Suppression of Reaginic Formation. J. Immunol. 109:34-89. 

29. Lahss, F., Weiler, E., and Hillmann, G. (1953) 
Myelom-Plasma-Proteine. Ill Mitteilung zur Immunochanie der 
Gaima-Myelom-Proteine . Z. ^3aturf orsch , Teil., 83:625-631. 

30. Jerne, N.K. (1974) Towards a Network Theory of the 
Immune System. Ann. Immunol. 125(c) :373-389 . 

31. Brient, B.W. and Nisonoff, A. (1970) Quantitative 
Investigations of Idiotypic Antibodies. IV. Inhibition by 
Specific Haptens of the Reaction of Anti-Hapten Antibody 
with its Anti-Idiotypic Antibody. J. Exp. Med. 
132:951-963. 

32. Claflin, J.L., and Davie, J.M. (1975) Specific 
Isolation and Characterization of Antibody Directed to 
Binding Site Antigenic Determinants. J,. Immunol. 114:70-75. 



143 



33. Brient, B.W. , Haimovich, J., and Nisonoff, A. (1971) 
Reaction of Anti-Idiotypic Antibody with the Hapten-binding 
Site of a I<^yeloma Protein. Proc. Natl. Acad. Sci. U.S.A. 
68:3136-3139. 

34. Briles, D.E., and Krause, R.M. (1974) ftouse 
Strain-Specific Idiotypy and Intrastrain Idiotypic Cross 
Reactions. J. Inmunol. 113:522-530. 

35. Sher, A., and Cohn, M. (1972) Effect of Haptens on 
the Reaction of Anti-Idiotype Antibody with a Mouse 
Anti-Phosfhorylcholine Plasmocytoma Protein. J. Iinnunol. 
109:176-178. 

36. Eichmann, K., Coutinho, A., and Melchers, F. (1977) 
Absolute Frequencies of Lipopolysaccharide-Reactive B Cells 
Producing ASA Idiotype in Unpriitied, Streptococcal A 
Carbohydrate-Primed Anti-A5A Idiotype-Sensitized and 
Anti-A5A Idiotype- Suppressed A/J Mice. J. Exp. Med. 
146:1436-1449. 

37. Bona, C, Mond, J.M., Stein, K.E., House, S., Liberman, 
R., and Paul, W.E. (1979) Immune Response to Levan. Ill 
The Capacity to Produce Anti-Inulin Antibodies and 
Cross-Reactive Idiotypes Appears Late in Ontogeny. J. 
Inmunol. 123:1484-1490. 

38. Marcu, K.B. (1982) Inmunoglobulin Heavy-Chain 
Constant-Region Genes. Cell 29:719-721. 

39. Kolb, H., and Bosm, M.J. (1977) Clones Producing 
Antibody of More than One Class. Inmunol. 33:461-469. 

40. Hopper, J.E., and Nisonoff, A. (1971) Individual 
Antigenic Specif ity of Inmunoglotxalins . Adv. Immunol. 
13:57-99. 

41. Capra, J.D., and Kohoe, J.M. (1975) Hyper variable 
Regions, Idiotypy, and the Antibody-Combining Site. Adv. 
Irtinunol . 20 : 1-40 . 

42. Eichmann, K. (1978) Expression and Function of 
Idiotypes on Lyn^ocytes. Adv. Irrmianol. 26:195-254. 

43. Janeway, C.A. (1980) Idiotypes, T-Cell Receptors and 
T-B Cooperation. Oontemp. Top. Iimunobiol . 9:171-203. 



144 



44. RcxJkey, L.S. (1980) Autoregulation of Immune 
Responses via Idiotype iSfetwork Interaction. Microbiol. Rev. 
44:631-659. 

45. Bankert, R.B., Bloor, A.G., and Jou, Y.H. (1982) 
Idiotypes: Their Presence on B and T Lymphocytes and their 
Role in the Regulation of the Immune Response. Vet. 
Intnunol. Immunopath. 3:147-184. 

46. Greene, M.I., bfelles, M.J., and Man- Sun, S. (1982) 
Regulation of Immunity to the Azobenzenearsonate Hapten. 
Adv. Inrnunol. 32:254-300. 

47. Lynch, R.G., Rohrer, J.W., Odermatt, B., Gebel, H.M., 
Autry, J.R. , and Hoover, R.G. (1979) Imnunoregulation of 
Murine Myeloma Cell Growth and Differentiation: A 
Monoclonal Model of B Cell Differentiation. Immunol. Rev. 
48:45-80. 

48. Cosenza, H., and Kohler, H. (1972) Specific 
Inhibition of Plaque Formation to Phosphorylcholine by 
Antibody Against Antibody. Science 176:1027-1029. 

49. Cosenza, H., and Kohler, H. (1972) Specific 
Suppression of the Antibody Response by Antibodies to 
Receptors. Proc. Natl. Acad. Sci. U.S.A. 69:2701-2705. 

50. Bankert, R.B., and Pressnan, D. (1976) 
Receptor-Blocking Factors Present in Iimune Serum Resembling 
Auto-Anti-Idiotype Antibody. J. Immunol. 117:457-462. 

51. Kelsoe, G., and Cerny, J. (1979) Reciprocal 
Expansions of Idiotypic and Anti-Idiotypic Clones Following 
Antigen Stimulation. Nature. 279:333-334. 

52. Geha, R.S. (1983) Presence of CirciiLating 
Anti-Idiotype Bearing Cells after Booster Immunization with 
Tetanus Toxoid (TD and Inhibition of Anti-TT Antibody 
Synthesis by Auto-Anti- Idiotypic Antibody. J. Iimiunol. 
130-1634-1639 . 

53. Dwyer, D.S., Bradley, R.J., Urquhart, C.K., and 
Kearney, J.F. (1983) Naturally Occurring Anti-Idiotypic 
Antibodies in Myasthenia Gravis Patients. l>feture 
301:611-614. 



54. Abdou, N.I., Wall, H., Lindsley, H.B,, Halsey, J.N. , 
and Suzuki, T. (1981) Network Theory in Autoimmunity. In 
Vitro Sijppression of Serum Anti-DNA. Antibody Binding DNA 
by Anti-Idiotypic Antibody in Systemic Lupus Erytharetosus . 
J. Clin. Invest. 67:1297-1304. 

55. Cunningham-Rundles, C. (1982) "Naturally Occuring 
Autologous Anti-Idiotypic Antibodies: Participation in 
IiTiiiune Complex Formation in Selective IgA Deficiency. J. 
Exp. lyted. 155:711-719. 

56. Schrater, A.F., Goidl, E.A., Thorbecke, G.J., and 
Siskind, G.W. (1979) Production of Auto-Anti- Idiotypic 
Antibody during the Normal Inmune Response to TNP-Ficoll. I 
Occurrence in AKR/J and BALB/C Mice of Hapten- Augmentable , 
Anti-TNP Plaque-Forming Cells and their Accelerated 
Appearance in Recipients of Immune Spleen Cells. J. Exp. 
Med. 150:138-153. 

57. Binion, S.B. , and Rodkey, L.S. (1982) Naturally 
Induced Auto-Anti-Idiotypic Antibodies. Induction by 
Identical Idiotypes in some Monbers of an Outbred Rabbit 
Family. J. Exp. Med. 156:860-872. 

58. Geczy, A.F., de Week, A.L., Geczy, C.L., and Toffler, 
0. (1978) Suppression of Reaginic Antibody Formation in 
Guinea Pigs by Anti-Idiotypic Antibodies. J. Allergy Clin. 
Iitmunol. 62:261-270. 

59. Blaser, K., Nakagawa, T., and de Week, A.L. (1981) 
Suppression of Anti-Hapten IgE and IgG Antibody Responses by 
Isologous Anti-Idiotypic Antibodies Against Purified 
Anti-Carrier (Ovalbumin) Antibodies in BALB/C Mice. J. 
Inrnunol. 126:1180-1184. 

60. Blaser, K., Nakagawa, T., and de Week, A.L. (1980) 
Sippression of the Benzylpenicelloyl (BPO) Specific IgE 
Formation with Isologous Anti-Idiotypic Antibodies in BALB/C 
Mice. J. Iirmunol. 125:24-30. 

61. Blaser, K., Geiser, M. , and de Week, A.L. (1979) 
Suppression of Phosphorylcholine Specific IgE Antibody 
Formation in Balb/C Mice by Isologous Anti-T15 Antiserum. 
Eur. J. Iimiunol. 9:1017-1020. 

62. Eichmann, K. , and Rajewsky, K. (1975) Induction of T 
and B Cell Immunity by Anti-Idiotypic Antibody. Eur. J. 
Iirmunol. 5:651-666. 



14 6 



63. Forni, L. , Coutinho, A., Kohler, G., and Jerne, N.K. 
(1980) IgM Antibody Induces the Production of Antibodies of 
the Sane Specificity. Proc. Natl. Acad Sci. U.S.A. 
77:1125-1128. 

64. Ivars, F., Holmberg, D., Forni, L. , Cazenave, P. A., and 
Coutinko, A. (1982) Antigen-Independent, IgiVt- Induced 
Antibody Responses: Requiranent for "Recurrent" Idiotypes. 
Eur. J. Irraiunol. 12:146-151. 

65. Sege, K. , and Peterson P.A. (1978) Ant i- Idiotypic 
Antibodies Against Anti-Vitamin A Transporting Protein React 
with Prealbumin. Nature. 271:167-168. 

66. Schreiber, A., Andre, C, Vray, B., and Strosberg, A.D. 
(1980) Anti-Alprenolol Anti-Idiotypic Antibodies Bind to 
B-Adrenergic Receptors and Modulate Catecholamine Sensitive 
Adenylate C^yclase. Proc. iSfetl. Acad. Sci. U.S.A. 
77:7385-7389. 

67. Schwartzman, R.M., Rockey, J.H., and Halliwell, R.E.W. 
(1971) Canine Reaginic Antibody. Characterization of tlie 
Spontaneous Anti-Ragweed and Induced Anti-Dinitroj^ienyl 
Reaginic Antibodies of the Atopic EOg. Clin. Exp. Immunol. 
9:549-569. 

68. Bradford, MM. (1976) A Rapid and Sensitive Method 
for the Quantification of Microgram Quantities of Protein 
Utilizing the Principle of Protein-Dye Binding. Anal 
Biochem. 72:248-254. 

69. Strejan, G., and Campbell, D.H. (1963) 
Hypersensitivity to Ascaris Antigen. IV Production of 
Honocytotropic Antibodies in the Rat. J. Immunol. 
101:628-637. 

70. Williams, C.A., and Chase, M.W. (1967) Methods in 
Immunology and Iirmunochgnistry . Vol. 1, Academic Press, 

York, New York. 

71. Imda, Y., Namoto, K., Yamada, H. , and Takeya, K. 
(1977) Relationships between IgE Antibody Production and 
other Immune Responses. 1. Ontogenic Development of IgE 
Antibody- Producing Capacity in Mice. Int. Arch Allergy 
Appl. Immunol. 53:50-55. 

72. Vaz, E.M., Vaz, N.M. , and Levine B.B. (1971) 
Persistent Formation of Reagins in Mice Injected with Low 
Doses of Ovalbumin. Iirmunol. 21:11-15. 



147 



73. Revoltella, R., and Ovary, Z. (1971) Reagenic 
Antibody Production in Different Mouse Strains. Immunol. 
17:45-54. 

74. Geha, R.S. (1981) Regulation of the Immune Response 
by Idiotype-Antiidiotype Interactions. N. Eng. J, Med. 
305:25-28. 

75. Jarrett, E.E.E., and Ferguson, A. (1974) Effect of 
T-Cell Depletion on the Potentiated Reagin Response. Nature 
250:420-422. 

76. Chiorazzi, N., Fox, D.A., and Katz, D.H, (1976) 
Hapten- Specific Enhancement of IgE Antibody Production by 
Low Doses of X- Irradiation and by Cyclophosphamide. J. 
IiTtnunol. 117:1629-1637. 

77. Nagel, J.E., White, C, Lin, M.S., and Firsnan, P. 
(1979) IgE Syntiiesis in Man, II. Comparison of Tetanus and 
Diptheria IgE Antibody in Allergic and Non-Allergic 
Children. J. Allergy Clin. Immunol. 63:308-314. 

78. Frich, O.L., and Brooks, D.L. (1983) Iirmunoglobulin E 
Antibodies to Pollens Augmented in Dogs by Virus Vaccines. 
Am. J. Vet. Res. 44:440-445. 

79. Katz, D, (1978) The Allergic Phenotype. 
Manifestation of "allergic breakthrough" and imbalance in 
nomal "Damping" of IgE Antibody Production. Immunol. Rev. 
41:77-108. 

80. Rowley, D.A. , Fitch, F.W. , Stuart, F.P., Kohler, H. , 
and Cosenza, H. (1973) Specific Sippression of Immune 
Responses. Science 181:1133-1141. 

81. Voisin, G.A. (1980) Role of Antibody Classes in the 
Regulatory Facilitation Reaction. Inmunol. Rev. 49:3-59. 

82. Hart, D.A. , Pawlak, L.L. and Nisonoff, A. (1973) 
Nature of Anti-Hapten Antibodies Arising after Immune 
Sippression of a Set of Cross-Reactive Idiotypic 
Specificities. Eur. J. Inmunol. 3:44-48. 

83. Jarrett, E.E.E., and Hall, E. (1983) IgE Suppression 
by Maternal IgG. Immunol. 48:49-59. 

84. Pawlak, L.L., Hart, D.A., and Nisonoff, A. (1973) 
Requiraiients for Prolonged Sippression of an Idiotypic 
Specificity in Adult Mice. J. Exp. j>fed. 137:1442-1458. 



14 8 



85. Claflin, J.L., and Davies, J.M. (1975) Clonal Nature 
of the Immune Response to Phosphoryl choline (PC). V. 
Cross- Idiotypic Specificity among Heavy Chains of Murine 
Anti-PC Antibodies and PC-Binding Myeloma Proteins. J. Exp. 
Med. 141:1073-1083. 

86. Karol, R.A., Reichlin, M. , and tfeble, R.W. (1977) 
Evolution of an Idiotypic Determinant: Anti-Val. J. Exp. 
Med. 146:435-444. 

87. Schwartz, M. , Norvick, D. , Givol, D., and Fuchs, S. 
(1978) Induction of Ant i- Idiotypic Antibodies by 
Iimiunization with Syngeneic Spleen Cells Educated with 
Acetylcholine Receptor. Nature 273:543-545. 

88. Eichmann, K. (1974) Idiotype Suppression: I 
Influence of the Effector Functions of Anti-Idiotypic 
Antibody on the Production of an Idiotype. Eur. J. Itmiunol. 
4:296-302. 

89. Nisonoff, A. and Lamoyi, E. (1981) Implications of 
the Presence of an Internal Image of the Antigen in 
Anti-Idiotypic Antibodies: Possible Application to Vaccine 
Production. Clin. Immunol, and Imi!iunopath . 21:397-406. 

90. Eisen, H.N. (1980) Iirmunology . Second Edition. 
Harper and Row Publishers, Hagerstown, Md. 

91. Theofilopoulos, A.N., and Dixon, F.J. (1979) The 
Biology and Detection of Immune Complexes. Adv. Immunol. 
28:39-220. 



BIOGRAPHICAL SKETCH 

My name is Tippy Schultz. I am a border oollie, spitz 
dog who was an abused puppy scheduled to be euthanized on 
the day I met my future cwner, Kevin Schultz. He rescued me 
frcxn tr^ fate and since he was in his first year of 
veterinary school at Purdue University, I became his living 
anatomical model to study dog topograj±iy. I learned that he 
had been born in Chicago in 1951, and, because his family 
moved alot, he attended many different grade and high 
schools before finally moving to Fort Wayne, Indiana, ^ere 
he graduated fron high school. We studied hard in 
veterinary school and I sent Kevin to work at B.F. Goodrich 
Tire Company during sumrtErs to keep me in food and shelter 
v^ich provided motivation for him to continue his education. 

We graduated in 1976 and since that time, I have 
gotten to do a great deal of traveling with him. In Dodge 
City, Kansas, while he was in practice, he met a very pretty 
and very nice voman. He saved her cat from dying after 
being hit by a car, and since she could also type, they got 
married and we all moved to Chicago so Kevin could practice 



14 9 



150 

for a year. Vie then hit the road for Philadelphia, Pa., 
vAiere Kevin studied comparative demiatologY at the 
University of Pennsylvania, and he put Jfency to wDrk typing. 
In September, 1979, we packed up and moved again, this time 
to Gainesville, Florida. Here both Kevin and Nancy attended 
the University of Florida and I didn't see much of either of 
thsn while they were busy getting their degrees. 

During our treks across the country, we have adopted 
two permanent house guests, Fanny and Pippin, ^o are OK for 
being just cats. In July, 1983, we will all be moving 
again, this time to Boston, Massachusetts. There Kevin will 
be a post-doctoral fellow in the Departm.ent of Pathology at 
Harvard iyiedical School which is fine with vcb as long as he 
continues to keep me v;arm and fed. 



I certify that I have read this study and that in my 
opinion it conforms to acceptable standards of scholarly 
presentation and is fully adequate, in scope and quality as 
a dissertation for the degree of Doctor of Philosophy. 



Richard E.W. Halliwell, Chairman 
Professor of Immunology and 
iVfedical Microbiology 



I certify that I have read this study and that in m/ 
opinion it conforms to acceptable standards of scholarly 
presentation and is fully adequate, in scope and quality as 
a dissertation for the degree of Doctor of Philosophy. 

George E. Gifford 
Professor of Immunology and 
Medical Microbiology 



I certify that I have read this study and that in my opinion 
it conforms to acceptable standards of scholarly 
presentation and is fully adequate, in scope and quality, as 
a dissertation for the degree of Doctor of Philosopl; 




Parker Smll/ Jr , 
Professor of Immunology and 
Medical Microbiology 



I certify that I have read this study and that in my opinion 
it conforms to acceptable standards of scholarly 
presentation and is fully adequate, in scope and quality as 
a dissertation for the degree of Doctor of Philosophy. 



Richard B. Crandall 
Professor of Immunology and 
Medical Microbiology 



I certify that I have read this study and that in rt^^ 
opinion it conforms to acceptable standards of scholarly 
presentation and is fully adequate, in scope and quality as 
a dissertation for the degree of Doctor of Philosophy. 




Michael D.P. Ayle 
Associate Professor of 
Iinnaunology and Medical 
Microbiology 



I certify that I have read this study and tiiat in opinion 
it conforms to acceptable standards of scholarly 
presentation and is fully adequate, in scope and quality as 
a dissertation for the degree of Doctor of Philosophy. 



Michael J. P. Lawman 
Assistant Professor of 
Veterinary Medicine 



This dissertation was submitted to tlie Graduate 
Faculty of the College of Medicine and to the Graduate 
Council, and was accepted as partial fvilfillment of the 
requirements for the degree of Doctor of Philosophy. 



August, 1983